diff --git a/.env.example b/.env.example index 06d43070c..e899d2adc 100644 --- a/.env.example +++ b/.env.example @@ -5,16 +5,13 @@ # ANTHROPIC_API_KEY=sk-ant-xxx # OPENAI_API_KEY=sk-xxx # GEMINI_API_KEY=xxx -# CEREBRAS_API_KEY=xxx - +# CLAUDE_CODE_OAUTH=xxx # ── Chat Channel ────────────────────────── # TELEGRAM_BOT_TOKEN=123456:ABC... # DISCORD_BOT_TOKEN=xxx -# LINE_CHANNEL_SECRET=xxx -# LINE_CHANNEL_ACCESS_TOKEN=xxx # ── Web Search (optional) ──────────────── # BRAVE_SEARCH_API_KEY=BSA... # ── Timezone ────────────────────────────── -TZ=Asia/Tokyo +TZ=Asia/Shanghai diff --git a/.github/workflows/pr.yml b/.github/workflows/pr.yml index be1c10c52..1e9a7919a 100644 --- a/.github/workflows/pr.yml +++ b/.github/workflows/pr.yml @@ -24,6 +24,25 @@ jobs: with: version: v2.10.1 + vuln_check: + name: Security Check + runs-on: ubuntu-latest + steps: + - name: Checkout + uses: actions/checkout@v6 + with: + persist-credentials: false + + - name: Setup Go + uses: actions/setup-go@v5 + with: + go-version-file: go.mod + + - name: Run Govulncheck + uses: golang/govulncheck-action@v1 + with: + go-package: ./... + test: name: Tests runs-on: ubuntu-latest diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml index 786c893ef..0edd29f22 100644 --- a/.github/workflows/release.yml +++ b/.github/workflows/release.yml @@ -17,6 +17,11 @@ on: required: false type: boolean default: false + upload_tos: + description: "Upload to Volcengine TOS" + required: false + type: boolean + default: true jobs: create-tag: @@ -100,3 +105,12 @@ jobs: gh release edit "${{ inputs.tag }}" \ --draft=${{ inputs.draft }} \ --prerelease=${{ inputs.prerelease }} + + upload-tos: + name: Upload to TOS + needs: release + if: ${{ inputs.upload_tos }} + uses: ./.github/workflows/upload-tos.yml + with: + tag: ${{ inputs.tag }} + secrets: inherit diff --git a/.github/workflows/upload-tos.yml b/.github/workflows/upload-tos.yml new file mode 100644 index 000000000..6d3916d53 --- /dev/null +++ b/.github/workflows/upload-tos.yml @@ -0,0 +1,49 @@ +name: Upload to Volcengine TOS + +on: + workflow_dispatch: + inputs: + tag: + description: "Release tag to download and upload (e.g. v0.2.0)" + required: true + type: string + workflow_call: + inputs: + tag: + description: "Release tag to download and upload" + required: true + type: string + +jobs: + upload-tos: + name: Upload to Volcengine TOS + runs-on: ubuntu-latest + steps: + - name: Download release assets + env: + GH_TOKEN: ${{ secrets.GITHUB_TOKEN }} + run: | + mkdir -p artifacts + gh release download "${{ inputs.tag }}" \ + --repo "${{ github.repository }}" \ + --dir artifacts \ + --pattern "*.tar.gz" \ + --pattern "*.zip" \ + --pattern "*.rpm" \ + --pattern "*.deb" + + - name: Upload to Volcengine TOS + env: + AWS_ACCESS_KEY_ID: ${{ secrets.VOLC_TOS_ACCESS_KEY }} + AWS_SECRET_ACCESS_KEY: ${{ secrets.VOLC_TOS_SECRET_KEY }} + AWS_DEFAULT_REGION: cn-beijing + run: | + aws configure set default.s3.addressing_style virtual + TOS_ENDPOINT="https://tos-s3-cn-beijing.volces.com" + # Upload to versioned directory + aws s3 sync artifacts/ "s3://picoclaw-downloads/${{ inputs.tag }}/" \ + --endpoint-url "$TOS_ENDPOINT" + # Upload to latest (overwrite) + aws s3 sync artifacts/ "s3://picoclaw-downloads/latest/" \ + --endpoint-url "$TOS_ENDPOINT" \ + --delete diff --git a/.gitignore b/.gitignore index 02ef18d1f..a52b8d25a 100644 --- a/.gitignore +++ b/.gitignore @@ -38,6 +38,9 @@ ralph/ .ralph/ tasks/ +# Plans +docs/plans/ + # Editors .vscode/ .idea/ diff --git a/LICENSE b/LICENSE index 410acae26..b38d9340d 100644 --- a/LICENSE +++ b/LICENSE @@ -19,7 +19,3 @@ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. - ---- - -PicoClaw is heavily inspired by and based on [nanobot](https://github.com/HKUDS/nanobot) by HKUDS. diff --git a/Makefile b/Makefile index afc76a6ad..8de98e984 100644 --- a/Makefile +++ b/Makefile @@ -18,6 +18,28 @@ LDFLAGS=-ldflags "-X $(INTERNAL).version=$(VERSION) -X $(INTERNAL).gitCommit=$(G GO?=CGO_ENABLED=0 go GOFLAGS?=-v -tags stdjson +# Patch MIPS LE ELF e_flags (offset 36) for NaN2008-only kernels (e.g. Ingenic X2600). +# +# Bytes (octal): \004 \024 \000 \160 → little-endian 0x70001404 +# 0x70000000 EF_MIPS_ARCH_32R2 MIPS32 Release 2 +# 0x00001000 EF_MIPS_ABI_O32 O32 ABI +# 0x00000400 EF_MIPS_NAN2008 IEEE 754-2008 NaN encoding +# 0x00000004 EF_MIPS_CPIC PIC calling sequence +# +# Go's GOMIPS=softfloat emits no FP instructions, so the NaN mode is irrelevant +# at runtime — this is purely an ELF metadata fix to satisfy the kernel's check. +# patchelf cannot modify e_flags; dd at a fixed offset is the most portable way. +# +# Ref: https://codebrowser.dev/linux/linux/arch/mips/include/asm/elf.h.html +define PATCH_MIPS_FLAGS + @if [ -f "$(1)" ]; then \ + printf '\004\024\000\160' | dd of=$(1) bs=1 seek=36 count=4 conv=notrunc 2>/dev/null || \ + { echo "Error: failed to patch MIPS e_flags for $(1)"; exit 1; }; \ + else \ + echo "Error: $(1) not found, cannot patch MIPS e_flags"; exit 1; \ + fi +endef + # Golangci-lint GOLANGCI_LINT?=golangci-lint @@ -50,6 +72,8 @@ ifeq ($(UNAME_S),Linux) ARCH=loong64 else ifeq ($(UNAME_M),riscv64) ARCH=riscv64 + else ifeq ($(UNAME_M),mipsel) + ARCH=mipsle else ARCH=$(UNAME_M) endif @@ -97,6 +121,8 @@ build-whatsapp-native: generate GOOS=linux GOARCH=arm64 $(GO) build -tags whatsapp_native $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-linux-arm64 ./$(CMD_DIR) GOOS=linux GOARCH=loong64 $(GO) build -tags whatsapp_native $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-linux-loong64 ./$(CMD_DIR) GOOS=linux GOARCH=riscv64 $(GO) build -tags whatsapp_native $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-linux-riscv64 ./$(CMD_DIR) + GOOS=linux GOARCH=mipsle GOMIPS=softfloat $(GO) build -tags whatsapp_native $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-linux-mipsle ./$(CMD_DIR) + $(call PATCH_MIPS_FLAGS,$(BUILD_DIR)/$(BINARY_NAME)-linux-mipsle) GOOS=darwin GOARCH=arm64 $(GO) build -tags whatsapp_native $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-darwin-arm64 ./$(CMD_DIR) GOOS=windows GOARCH=amd64 $(GO) build -tags whatsapp_native $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-windows-amd64.exe ./$(CMD_DIR) ## @$(GO) build $(GOFLAGS) -tags whatsapp_native $(LDFLAGS) -o $(BINARY_PATH) ./$(CMD_DIR) @@ -117,6 +143,14 @@ build-linux-arm64: generate GOOS=linux GOARCH=arm64 $(GO) build $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-linux-arm64 ./$(CMD_DIR) @echo "Build complete: $(BUILD_DIR)/$(BINARY_NAME)-linux-arm64" +## build-linux-mipsle: Build for Linux MIPS32 LE +build-linux-mipsle: generate + @echo "Building for linux/mipsle (softfloat)..." + @mkdir -p $(BUILD_DIR) + GOOS=linux GOARCH=mipsle GOMIPS=softfloat $(GO) build $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-linux-mipsle ./$(CMD_DIR) + $(call PATCH_MIPS_FLAGS,$(BUILD_DIR)/$(BINARY_NAME)-linux-mipsle) + @echo "Build complete: $(BUILD_DIR)/$(BINARY_NAME)-linux-mipsle" + ## build-pi-zero: Build for Raspberry Pi Zero 2 W (32-bit and 64-bit) build-pi-zero: build-linux-arm build-linux-arm64 @echo "Pi Zero 2 W builds: $(BUILD_DIR)/$(BINARY_NAME)-linux-arm (32-bit), $(BUILD_DIR)/$(BINARY_NAME)-linux-arm64 (64-bit)" @@ -130,6 +164,8 @@ build-all: generate GOOS=linux GOARCH=arm64 $(GO) build $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-linux-arm64 ./$(CMD_DIR) GOOS=linux GOARCH=loong64 $(GO) build $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-linux-loong64 ./$(CMD_DIR) GOOS=linux GOARCH=riscv64 $(GO) build $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-linux-riscv64 ./$(CMD_DIR) + GOOS=linux GOARCH=mipsle GOMIPS=softfloat $(GO) build $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-linux-mipsle ./$(CMD_DIR) + $(call PATCH_MIPS_FLAGS,$(BUILD_DIR)/$(BINARY_NAME)-linux-mipsle) GOOS=linux GOARCH=arm GOARM=7 $(GO) build $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-linux-armv7 ./$(CMD_DIR) GOOS=darwin GOARCH=arm64 $(GO) build $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-darwin-arm64 ./$(CMD_DIR) GOOS=windows GOARCH=amd64 $(GO) build $(LDFLAGS) -o $(BUILD_DIR)/$(BINARY_NAME)-windows-amd64.exe ./$(CMD_DIR) diff --git a/README.fr.md b/README.fr.md index e537fc13a..08a1926b6 100644 --- a/README.fr.md +++ b/README.fr.md @@ -7,7 +7,7 @@

Go - Hardware + Hardware License
Website @@ -65,7 +65,7 @@ ⚡️ **Démarrage Éclair** : Temps de démarrage 400X plus rapide, boot en 1 seconde même sur un cœur unique à 0,6 GHz. -🌍 **Véritable Portabilité** : Un seul binaire autonome pour RISC-V, ARM et x86. Un clic et c'est parti ! +🌍 **Véritable Portabilité** : Un seul binaire autonome pour RISC-V, ARM, MIPS et x86. Un clic et c'est parti ! 🤖 **Auto-Construit par l'IA** : Implémentation native en Go de manière autonome — 95% du cœur généré par l'Agent avec affinement humain dans la boucle. @@ -827,7 +827,7 @@ Le sous-agent a accès aux outils (message, web_search, etc.) et peut communique ### Fournisseurs > [!NOTE] -> Groq fournit la transcription vocale gratuite via Whisper. Si configuré, les messages vocaux Telegram seront automatiquement transcrits. +> Groq fournit la transcription vocale gratuite via Whisper. Si configuré, les messages audio de n'importe quel canal seront automatiquement transcrits au niveau de l'agent. | Fournisseur | Utilisation | Obtenir une Clé API | | ------------------------ | ---------------------------------------- | ------------------------------------------------------ | diff --git a/README.ja.md b/README.ja.md index 20ad5033b..c4c5b27a0 100644 --- a/README.ja.md +++ b/README.ja.md @@ -8,7 +8,7 @@

Go -Hardware +Hardware License

@@ -49,7 +49,7 @@ ⚡️ **超高速**: 起動時間 400 倍高速、0.6GHz シングルコアでも 1 秒で起動。 -🌍 **真のポータビリティ**: RISC-V、ARM、x86 対応の単一バイナリ。ワンクリックで Go！ +🌍 **真のポータビリティ**: RISC-V、ARM、MIPS、x86 対応の単一バイナリ。ワンクリックで Go！ 🤖 **AI ブートストラップ**: 自律的な Go ネイティブ実装 — コアの 95% が AI 生成、人間によるレビュー付き。 @@ -785,7 +785,7 @@ HEARTBEAT_OK 応答 ユーザーが直接結果を受け取る ### プロバイダー > [!NOTE] -> Groq は Whisper による無料の音声文字起こしを提供しています。設定すると、Telegram の音声メッセージが自動的に文字起こしされます。 +> Groq は Whisper による無料の音声文字起こしを提供しています。設定すると、あらゆるチャンネルからの音声メッセージがエージェントレベルで自動的に文字起こしされます。 | プロバイダー | 用途 | API キー取得先 | | --- | --- | --- | diff --git a/README.md b/README.md index c5b38e222..db127a85f 100644 --- a/README.md +++ b/README.md @@ -7,7 +7,7 @@

Go - Hardware + Hardware License
Website @@ -54,7 +54,7 @@ ## 📢 News -2026-02-16 🎉 PicoClaw hit 12K stars in one week! Thank you all for your support! PicoClaw is growing faster than we ever imagined. Given the high volume of PRs, we urgently need community maintainers. Our volunteer roles and roadmap are officially posted [here](docs/ROADMAP.md) —we can’t wait to have you on board! +2026-02-16 🎉 PicoClaw hit 12K stars in one week! Thank you all for your support! PicoClaw is growing faster than we ever imagined. Given the high volume of PRs, we urgently need community maintainers. Our volunteer roles and roadmap are officially posted [here](ROADMAP.md) —we can’t wait to have you on board! 2026-02-13 🎉 PicoClaw hit 5000 stars in 4days! Thank you for the community! There are so many PRs & issues coming in (during Chinese New Year holidays), we are finalizing the Project Roadmap and setting up the Developer Group to accelerate PicoClaw's development. 🚀 Call to Action: Please submit your feature requests in GitHub Discussions. We will review and prioritize them during our upcoming weekly meeting. @@ -69,7 +69,7 @@ ⚡️ **Lightning Fast**: 400X Faster startup time, boot in 1 second even in 0.6GHz single core. -🌍 **True Portability**: Single self-contained binary across RISC-V, ARM, and x86, One-click to Go! +🌍 **True Portability**: Single self-contained binary across RISC-V, ARM, MIPS, and x86, One-click to Go! 🤖 **AI-Bootstrapped**: Autonomous Go-native implementation — 95% Agent-generated core with human-in-the-loop refinement. @@ -216,7 +216,7 @@ docker compose -f docker/docker-compose.yml --profile gateway up -d > [!TIP] > Set your API key in `~/.picoclaw/config.json`. > Get API keys: [OpenRouter](https://openrouter.ai/keys) (LLM) · [Zhipu](https://open.bigmodel.cn/usercenter/proj-mgmt/apikeys) (LLM) -> Web Search is **optional** - get free [Tavily API](https://tavily.com) (1000 free queries/month) or [Brave Search API](https://brave.com/search/api) (2000 free queries/month) or use built-in auto fallback. +> Web Search is **optional** - get free [Tavily API](https://tavily.com) (1000 free queries/month), [SearXNG](https://github.com/searxng/searxng) (free, self-hosted) or [Brave Search API](https://brave.com/search/api) (2000 free queries/month) or use built-in auto fallback. **1. Initialize** @@ -265,6 +265,16 @@ picoclaw onboard "duckduckgo": { "enabled": true, "max_results": 5 + }, + "perplexity": { + "enabled": false, + "api_key": "YOUR_PERPLEXITY_API_KEY", + "max_results": 5 + }, + "searxng": { + "enabled": false, + "base_url": "http://your-searxng-instance:8888", + "max_results": 5 } } } @@ -277,7 +287,12 @@ picoclaw onboard **3. Get API Keys** * **LLM Provider**: [OpenRouter](https://openrouter.ai/keys) · [Zhipu](https://open.bigmodel.cn/usercenter/proj-mgmt/apikeys) · [Anthropic](https://console.anthropic.com) · [OpenAI](https://platform.openai.com) · [Gemini](https://aistudio.google.com/api-keys) -* **Web Search** (optional): [Tavily](https://tavily.com) - Optimized for AI Agents (1000 requests/month) · [Brave Search](https://brave.com/search/api) - Free tier available (2000 requests/month) +* **Web Search** (optional): + * [Brave Search](https://brave.com/search/api) - Paid ($5/1000 queries, ~$5-6/month) + * [Perplexity](https://www.perplexity.ai) - AI-powered search with chat interface + * [SearXNG](https://github.com/searxng/searxng) - Self-hosted metasearch engine (free, no API key needed) + * [Tavily](https://tavily.com) - Optimized for AI Agents (1000 requests/month) + * DuckDuckGo - Built-in fallback (no API key required) > **Note**: See `config.example.json` for a complete configuration template. @@ -338,6 +353,13 @@ Talk to your picoclaw through Telegram, Discord, WhatsApp, DingTalk, LINE, or We picoclaw gateway ``` +**4. Telegram command menu (auto-registered at startup)** + +PicoClaw now keeps command definitions in one shared registry. On startup, Telegram will automatically register supported bot commands (for example `/start`, `/help`, `/show`, `/list`) so command menu and runtime behavior stay in sync. +Telegram command menu registration remains channel-local discovery UX; generic command execution is handled centrally in the agent loop via the commands executor. + +If command registration fails (network/API transient errors), the channel still starts and PicoClaw retries registration in the background. +

@@ -735,6 +757,12 @@ For advanced/test setups, you can override the builtin skills root with: export PICOCLAW_BUILTIN_SKILLS=/path/to/skills ``` +### Unified Command Execution Policy + +- Generic slash commands are executed through a single path in `pkg/agent/loop.go` via `commands.Executor`. +- Channel adapters no longer consume generic commands locally; they forward inbound text to the bus/agent path. Telegram still auto-registers supported commands at startup. +- Unknown slash command (for example `/foo`) passes through to normal LLM processing. +- Registered but unsupported command on the current channel (for example `/show` on WhatsApp) returns an explicit user-facing error and stops further processing. ### 🔒 Security Sandbox PicoClaw runs in a sandboxed environment by default. The agent can only access files and execute commands within the configured workspace. @@ -911,7 +939,7 @@ The subagent has access to tools (message, web_search, etc.) and can communicate ### Providers > [!NOTE] -> Groq provides free voice transcription via Whisper. If configured, Telegram voice messages will be automatically transcribed. +> Groq provides free voice transcription via Whisper. If configured, audio messages from any channel will be automatically transcribed at the agent level. | Provider | Purpose | Get API Key | | -------------------------- | --------------------------------------- | -------------------------------------------------------------------- | @@ -1190,6 +1218,10 @@ picoclaw agent -m "Hello" "model": "anthropic/claude-opus-4-5" } }, + "session": { + "dm_scope": "per-channel-peer", + "backlog_limit": 20 + }, "providers": { "openrouter": { "api_key": "sk-or-v1-xxx" @@ -1241,6 +1273,16 @@ picoclaw agent -m "Hello" "duckduckgo": { "enabled": true, "max_results": 5 + }, + "perplexity": { + "enabled": false, + "api_key": "", + "max_results": 5 + }, + "searxng": { + "enabled": false, + "base_url": "http://localhost:8888", + "max_results": 5 } }, "cron": { @@ -1298,10 +1340,69 @@ discord: This is normal if you haven't configured a search API key yet. PicoClaw will provide helpful links for manual searching. -To enable web search: +#### Search Provider Priority -1. **Option 1 (Recommended)**: Get a free API key at [https://brave.com/search/api](https://brave.com/search/api) (2000 free queries/month) for the best results. -2. **Option 2 (No Credit Card)**: If you don't have a key, we automatically fall back to **DuckDuckGo** (no key required). +PicoClaw automatically selects the best available search provider in this order: +1. **Perplexity** (if enabled and API key configured) - AI-powered search with citations +2. **Brave Search** (if enabled and API key configured) - Privacy-focused paid API ($5/1000 queries) +3. **SearXNG** (if enabled and base_url configured) - Self-hosted metasearch aggregating 70+ engines (free) +4. **DuckDuckGo** (if enabled, default fallback) - No API key required (free) + +#### Web Search Configuration Options + +**Option 1 (Best Results)**: Perplexity AI Search +```json +{ + "tools": { + "web": { + "perplexity": { + "enabled": true, + "api_key": "YOUR_PERPLEXITY_API_KEY", + "max_results": 5 + } + } + } +} +``` + +**Option 2 (Paid API)**: Get an API key at [https://brave.com/search/api](https://brave.com/search/api) ($5/1000 queries, ~$5-6/month) +```json +{ + "tools": { + "web": { + "brave": { + "enabled": true, + "api_key": "YOUR_BRAVE_API_KEY", + "max_results": 5 + } + } + } +} +``` + +**Option 3 (Self-Hosted)**: Deploy your own [SearXNG](https://github.com/searxng/searxng) instance +```json +{ + "tools": { + "web": { + "searxng": { + "enabled": true, + "base_url": "http://your-server:8888", + "max_results": 5 + } + } + } +} +``` + +Benefits of SearXNG: +- **Zero cost**: No API fees or rate limits +- **Privacy-focused**: Self-hosted, no tracking +- **Aggregate results**: Queries 70+ search engines simultaneously +- **Perfect for cloud VMs**: Solves datacenter IP blocking issues (Oracle Cloud, GCP, AWS, Azure) +- **No API key needed**: Just deploy and configure the base URL + +**Option 4 (No Setup Required)**: DuckDuckGo is enabled by default as fallback (no API key needed) Add the key to `~/.picoclaw/config.json` if using Brave: @@ -1317,6 +1418,16 @@ Add the key to `~/.picoclaw/config.json` if using Brave: "duckduckgo": { "enabled": true, "max_results": 5 + }, + "perplexity": { + "enabled": false, + "api_key": "YOUR_PERPLEXITY_API_KEY", + "max_results": 5 + }, + "searxng": { + "enabled": false, + "base_url": "http://your-searxng-instance:8888", + "max_results": 5 } } } @@ -1335,10 +1446,11 @@ This happens when another instance of the bot is running. Make sure only one `pi ## 📝 API Key Comparison -| Service | Free Tier | Use Case | -| ---------------- | ------------------- | ------------------------------------- | -| **OpenRouter** | 200K tokens/month | Multiple models (Claude, GPT-4, etc.) | -| **Zhipu** | 200K tokens/month | Best for Chinese users | -| **Brave Search** | 2000 queries/month | Web search functionality | -| **Groq** | Free tier available | Fast inference (Llama, Mixtral) | -| **Cerebras** | Free tier available | Fast inference (Llama, Qwen, etc.) | +| Service | Free Tier | Use Case | +| ---------------- | ------------------------ | ------------------------------------- | +| **OpenRouter** | 200K tokens/month | Multiple models (Claude, GPT-4, etc.) | +| **Zhipu** | 200K tokens/month | Best for Chinese users | +| **Brave Search** | Paid ($5/1000 queries) | Web search functionality | +| **SearXNG** | Unlimited (self-hosted) | Privacy-focused metasearch (70+ engines) | +| **Groq** | Free tier available | Fast inference (Llama, Mixtral) | +| **Cerebras** | Free tier available | Fast inference (Llama, Qwen, etc.) | diff --git a/README.pt-br.md b/README.pt-br.md index bfe655770..5f37ba457 100644 --- a/README.pt-br.md +++ b/README.pt-br.md @@ -7,7 +7,7 @@

Go - Hardware + Hardware License
Website @@ -66,7 +66,7 @@ ⚡️ **Inicialização Relámpago**: Tempo de inicialização 400X mais rápido, boot em 1 segundo mesmo em CPU single-core de 0.6GHz. -🌍 **Portabilidade Real**: Um único binário auto-contido para RISC-V, ARM e x86. Um clique e já era! +🌍 **Portabilidade Real**: Um único binário auto-contido para RISC-V, ARM, MIPS e x86. Um clique e já era! 🤖 **Auto-Construído por IA**: Implementação nativa em Go de forma autônoma — 95% do núcleo gerado pelo Agente com refinamento humano no loop. @@ -823,7 +823,7 @@ O subagente tem acesso às ferramentas (message, web_search, etc.) e pode se com ### Provedores > [!NOTE] -> O Groq fornece transcrição de voz gratuita via Whisper. Se configurado, mensagens de voz do Telegram serão automaticamente transcritas. +> O Groq fornece transcrição de voz gratuita via Whisper. Se configurado, mensagens de áudio de qualquer canal serão automaticamente transcritas no nível do agente. | Provedor | Finalidade | Obter API Key | | --- | --- | --- | diff --git a/README.vi.md b/README.vi.md index b30659614..92c6ecbae 100644 --- a/README.vi.md +++ b/README.vi.md @@ -7,7 +7,7 @@

Go - Hardware + Hardware License
Website @@ -65,7 +65,7 @@ ⚡️ **Khởi động siêu nhanh**: Nhanh gấp 400 lần, khởi động trong 1 giây ngay cả trên CPU đơn nhân 0.6GHz. -🌍 **Di động thực sự**: Một file binary duy nhất chạy trên RISC-V, ARM và x86. Một click là chạy! +🌍 **Di động thực sự**: Một file binary duy nhất chạy trên RISC-V, ARM, MIPS và x86. Một click là chạy! 🤖 **AI tự xây dựng**: Triển khai Go-native tự động — 95% mã nguồn cốt lõi được Agent tạo ra, với sự tinh chỉnh của con người. @@ -795,7 +795,7 @@ Subagent có quyền truy cập các công cụ (message, web_search, v.v.) và ### Nhà cung cấp (Providers) > [!NOTE] -> Groq cung cấp dịch vụ chuyển giọng nói thành văn bản miễn phí qua Whisper. Nếu đã cấu hình Groq, tin nhắn thoại trên Telegram sẽ được tự động chuyển thành văn bản. +> Groq cung cấp dịch vụ chuyển giọng nói thành văn bản miễn phí qua Whisper. Nếu đã cấu hình Groq, tin nhắn âm thanh từ bất kỳ kênh nào sẽ được tự động chuyển thành văn bản ở cấp độ agent. | Nhà cung cấp | Mục đích | Lấy API Key | | --- | --- | --- | diff --git a/README.zh.md b/README.zh.md index db96ba555..d42b3cbb8 100644 --- a/README.zh.md +++ b/README.zh.md @@ -7,7 +7,7 @@

Go - Hardware + Hardware License
Website @@ -67,7 +67,7 @@ ⚡️ **闪电启动**: 启动速度快 400 倍，即使在 0.6GHz 单核处理器上也能在 1 秒内启动。 -🌍 **真正可移植**: 跨 RISC-V、ARM 和 x86 架构的单二进制文件，一键运行！ +🌍 **真正可移植**: 跨 RISC-V、ARM、MIPS 和 x86 架构的单二进制文件，一键运行！ 🤖 **AI 自举**: 纯 Go 语言原生实现 — 95% 的核心代码由 Agent 生成，并经由“人机回环 (Human-in-the-loop)”微调。 @@ -307,6 +307,13 @@ PicoClaw 支持多种聊天平台，使您的 Agent 能够连接到任何地方 | **OneBot** | ⭐⭐ 中等 | 兼容 NapCat/Go-CQHTTP，社区生态丰富 | [查看文档](docs/channels/onebot/README.zh.md) | | **MaixCam** | ⭐ 简单 | 专为 AI 摄像头设计的硬件集成通道 | [查看文档](docs/channels/maixcam/README.zh.md) | +### Telegram 命令注册（启动时自动同步） + +PicoClaw 现在使用统一的命令定义来源。启动时会自动将 Telegram 支持的命令（例如 `/start`、`/help`、`/show`、`/list`）注册到 Bot 命令菜单，确保菜单展示与实际行为一致。 +Telegram 侧保留的是命令菜单注册能力；通用命令的实际执行统一走 Agent Loop 中的 commands executor。 + +如果注册因网络或 API 短暂异常失败，不会阻塞 channel 启动；系统会在后台自动重试。 + ## ClawdChat 加入 Agent 社交网络 只需通过 CLI 或任何集成的聊天应用发送一条消息，即可将 PicoClaw 连接到 Agent 社交网络。 @@ -376,6 +383,12 @@ PicoClaw 将数据存储在您配置的工作区中（默认：`~/.picoclaw/work export PICOCLAW_BUILTIN_SKILLS=/path/to/skills ``` +### 统一命令执行策略 + +- 通用斜杠命令通过 `pkg/agent/loop.go` 中的 `commands.Executor` 统一执行。 +- Channel 适配器不再在本地消费通用命令；它们只负责把入站文本转发到 bus/agent 路径。Telegram 仍会在启动时自动注册其支持的命令菜单。 +- 未注册的斜杠命令（例如 `/foo`）会透传给 LLM 按普通输入处理。 +- 已注册但当前 channel 不支持的命令（例如 WhatsApp 上的 `/show`）会返回明确的用户可见错误，并停止后续处理。 ### 心跳 / 周期性任务 (Heartbeat) PicoClaw 可以自动执行周期性任务。在工作区创建 `HEARTBEAT.md` 文件： @@ -459,7 +472,7 @@ Agent 读取 HEARTBEAT.md ### 提供商 (Providers) > [!NOTE] -> Groq 通过 Whisper 提供免费的语音转录。如果配置了 Groq，Telegram 语音消息将被自动转录为文字。 +> Groq 通过 Whisper 提供免费的语音转录。如果配置了 Groq，任意渠道的音频消息都将在 Agent 层面自动转录为文字。 | 提供商 | 用途 | 获取 API Key | | -------------------- | ---------------------------- | -------------------------------------------------------------------- | @@ -715,6 +728,10 @@ picoclaw agent -m "你好" "model": "anthropic/claude-opus-4-5" } }, + "session": { + "dm_scope": "per-channel-peer", + "backlog_limit": 20 + }, "providers": { "openrouter": { "api_key": "sk-or-v1-xxx" diff --git a/assets/wechat.png b/assets/wechat.png index 1c0b88295..32998c122 100644 Binary files a/assets/wechat.png and b/assets/wechat.png differ diff --git a/cmd/picoclaw-launcher-tui/internal/ui/model.go b/cmd/picoclaw-launcher-tui/internal/ui/model.go index ba91f5b09..304b4efa7 100644 --- a/cmd/picoclaw-launcher-tui/internal/ui/model.go +++ b/cmd/picoclaw-launcher-tui/internal/ui/model.go @@ -335,7 +335,11 @@ func (s *appState) testModel(model *picoclawconfig.ModelConfig) { s.showMessage("Test OK", resp.Status) return } - body, _ := io.ReadAll(io.LimitReader(resp.Body, 2048)) + body, err := io.ReadAll(io.LimitReader(resp.Body, 2048)) + if err != nil { + s.showMessage("Test failed", fmt.Sprintf("failed to read response: %v", err)) + return + } s.showMessage( "Test failed", fmt.Sprintf("%s: %s", resp.Status, strings.TrimSpace(string(body))), diff --git a/cmd/picoclaw-launcher/internal/server/auth_handlers.go b/cmd/picoclaw-launcher/internal/server/auth_handlers.go index 1e9b8be0a..3b48f9739 100644 --- a/cmd/picoclaw-launcher/internal/server/auth_handlers.go +++ b/cmd/picoclaw-launcher/internal/server/auth_handlers.go @@ -297,7 +297,10 @@ func fetchGoogleUserEmail(accessToken string) (string, error) { } defer resp.Body.Close() - body, _ := io.ReadAll(resp.Body) + body, err := io.ReadAll(resp.Body) + if err != nil { + return "", fmt.Errorf("reading userinfo response: %w", err) + } if resp.StatusCode != http.StatusOK { return "", fmt.Errorf("userinfo request failed: %s", string(body)) } diff --git a/cmd/picoclaw-launcher/internal/ui/index.html b/cmd/picoclaw-launcher/internal/ui/index.html index 93893fd75..d84fd4e6e 100644 --- a/cmd/picoclaw-launcher/internal/ui/index.html +++ b/cmd/picoclaw-launcher/internal/ui/index.html @@ -1392,9 +1392,7 @@ function saveModelFromModal() { saveConfig().then(renderModels); } -document.getElementById('modelModal').addEventListener('click', function(e) { - if (e.target === this) closeModelModal(); -}); + // ── Channel Forms ─────────────────────────────────── function renderChannelForm(chKey) { diff --git a/cmd/picoclaw/internal/auth/helpers.go b/cmd/picoclaw/internal/auth/helpers.go index 633ce8740..a0a229167 100644 --- a/cmd/picoclaw/internal/auth/helpers.go +++ b/cmd/picoclaw/internal/auth/helpers.go @@ -1,6 +1,7 @@ package auth import ( + "bufio" "encoding/json" "fmt" "io" @@ -15,14 +16,17 @@ import ( "github.com/sipeed/picoclaw/pkg/providers" ) -const supportedProvidersMsg = "supported providers: openai, anthropic, google-antigravity" +const ( + supportedProvidersMsg = "supported providers: openai, anthropic, google-antigravity" + defaultAnthropicModel = "claude-sonnet-4.6" +) -func authLoginCmd(provider string, useDeviceCode bool) error { +func authLoginCmd(provider string, useDeviceCode bool, useOauth bool) error { switch provider { case "openai": return authLoginOpenAI(useDeviceCode) case "anthropic": - return authLoginPasteToken(provider) + return authLoginAnthropic(useOauth) case "google-antigravity", "antigravity": return authLoginGoogleAntigravity() default: @@ -163,6 +167,81 @@ func authLoginGoogleAntigravity() error { return nil } +func authLoginAnthropic(useOauth bool) error { + if useOauth { + return authLoginAnthropicSetupToken() + } + + fmt.Println("Anthropic login method:") + fmt.Println(" 1) Setup token (from `claude setup-token`) (Recommended)") + fmt.Println(" 2) API key (from console.anthropic.com)") + + scanner := bufio.NewScanner(os.Stdin) + for { + fmt.Print("Choose [1]: ") + choice := "1" + if scanner.Scan() { + text := strings.TrimSpace(scanner.Text()) + if text != "" { + choice = text + } + } + + switch choice { + case "1": + return authLoginAnthropicSetupToken() + case "2": + return authLoginPasteToken("anthropic") + default: + fmt.Printf("Invalid choice: %s. Please enter 1 or 2.\n", choice) + } + } +} + +func authLoginAnthropicSetupToken() error { + cred, err := auth.LoginSetupToken(os.Stdin) + if err != nil { + return fmt.Errorf("login failed: %w", err) + } + + if err = auth.SetCredential("anthropic", cred); err != nil { + return fmt.Errorf("failed to save credentials: %w", err) + } + + appCfg, err := internal.LoadConfig() + if err == nil { + appCfg.Providers.Anthropic.AuthMethod = "oauth" + + found := false + for i := range appCfg.ModelList { + if isAnthropicModel(appCfg.ModelList[i].Model) { + appCfg.ModelList[i].AuthMethod = "oauth" + found = true + break + } + } + if !found { + appCfg.ModelList = append(appCfg.ModelList, config.ModelConfig{ + ModelName: defaultAnthropicModel, + Model: "anthropic/" + defaultAnthropicModel, + AuthMethod: "oauth", + }) + // Only set default model if user has no default configured yet + if appCfg.Agents.Defaults.GetModelName() == "" { + appCfg.Agents.Defaults.ModelName = defaultAnthropicModel + } + } + + if err := config.SaveConfig(internal.GetConfigPath(), appCfg); err != nil { + return fmt.Errorf("could not update config: %w", err) + } + } + + fmt.Println("Setup token saved for Anthropic!") + + return nil +} + func fetchGoogleUserEmail(accessToken string) (string, error) { req, err := http.NewRequest("GET", "https://www.googleapis.com/oauth2/v2/userinfo", nil) if err != nil { @@ -177,7 +256,10 @@ func fetchGoogleUserEmail(accessToken string) (string, error) { } defer resp.Body.Close() - body, _ := io.ReadAll(resp.Body) + body, err := io.ReadAll(resp.Body) + if err != nil { + return "", fmt.Errorf("reading userinfo response: %w", err) + } if resp.StatusCode != http.StatusOK { return "", fmt.Errorf("userinfo request failed: %s", string(body)) } @@ -217,13 +299,12 @@ func authLoginPasteToken(provider string) error { } if !found { appCfg.ModelList = append(appCfg.ModelList, config.ModelConfig{ - ModelName: "claude-sonnet-4.6", - Model: "anthropic/claude-sonnet-4.6", + ModelName: defaultAnthropicModel, + Model: "anthropic/" + defaultAnthropicModel, AuthMethod: "token", }) + appCfg.Agents.Defaults.ModelName = defaultAnthropicModel } - // Update default model - appCfg.Agents.Defaults.ModelName = "claude-sonnet-4.6" case "openai": appCfg.Providers.OpenAI.AuthMethod = "token" // Update ModelList @@ -360,6 +441,16 @@ func authStatusCmd() error { if !cred.ExpiresAt.IsZero() { fmt.Printf(" Expires: %s\n", cred.ExpiresAt.Format("2006-01-02 15:04")) } + + if provider == "anthropic" && cred.AuthMethod == "oauth" { + usage, err := auth.FetchAnthropicUsage(cred.AccessToken) + if err != nil { + fmt.Printf(" Usage: unavailable (%v)\n", err) + } else { + fmt.Printf(" Usage (5h): %.1f%%\n", usage.FiveHourUtilization*100) + fmt.Printf(" Usage (7d): %.1f%%\n", usage.SevenDayUtilization*100) + } + } } return nil diff --git a/cmd/picoclaw/internal/auth/login.go b/cmd/picoclaw/internal/auth/login.go index 9a6d28d2f..afbe098aa 100644 --- a/cmd/picoclaw/internal/auth/login.go +++ b/cmd/picoclaw/internal/auth/login.go @@ -6,6 +6,7 @@ func newLoginCommand() *cobra.Command { var ( provider string useDeviceCode bool + useOauth bool ) cmd := &cobra.Command{ @@ -13,12 +14,16 @@ func newLoginCommand() *cobra.Command { Short: "Login via OAuth or paste token", Args: cobra.NoArgs, RunE: func(cmd *cobra.Command, _ []string) error { - return authLoginCmd(provider, useDeviceCode) + return authLoginCmd(provider, useDeviceCode, useOauth) }, } cmd.Flags().StringVarP(&provider, "provider", "p", "", "Provider to login with (openai, anthropic)") cmd.Flags().BoolVar(&useDeviceCode, "device-code", false, "Use device code flow (for headless environments)") + cmd.Flags().BoolVar( + &useOauth, "setup-token", false, + "Use setup-token flow for Anthropic (from `claude setup-token`)", + ) _ = cmd.MarkFlagRequired("provider") return cmd diff --git a/cmd/picoclaw/internal/gateway/helpers.go b/cmd/picoclaw/internal/gateway/helpers.go index 747f7d44e..174f5db62 100644 --- a/cmd/picoclaw/internal/gateway/helpers.go +++ b/cmd/picoclaw/internal/gateway/helpers.go @@ -36,6 +36,7 @@ import ( "github.com/sipeed/picoclaw/pkg/providers" "github.com/sipeed/picoclaw/pkg/state" "github.com/sipeed/picoclaw/pkg/tools" + "github.com/sipeed/picoclaw/pkg/voice" ) func gatewayCmd(debug bool) error { @@ -134,6 +135,12 @@ func gatewayCmd(debug bool) error { agentLoop.SetChannelManager(channelManager) agentLoop.SetMediaStore(mediaStore) + // Wire up voice transcription if a supported provider is configured. + if transcriber := voice.DetectTranscriber(cfg); transcriber != nil { + agentLoop.SetTranscriber(transcriber) + logger.InfoCF("voice", "Transcription enabled (agent-level)", map[string]any{"provider": transcriber.Name()}) + } + enabledChannels := channelManager.GetEnabledChannels() if len(enabledChannels) > 0 { fmt.Printf("✓ Channels enabled: %s\n", enabledChannels) @@ -223,19 +230,25 @@ func setupCronTool( // Create cron service cronService := cron.NewCronService(cronStorePath, nil) - // Create and register CronTool - cronTool, err := tools.NewCronTool(cronService, agentLoop, msgBus, workspace, restrict, execTimeout, cfg) - if err != nil { - log.Fatalf("Critical error during CronTool initialization: %v", err) + // Create and register CronTool if enabled + var cronTool *tools.CronTool + if cfg.Tools.IsToolEnabled("cron") { + var err error + cronTool, err = tools.NewCronTool(cronService, agentLoop, msgBus, workspace, restrict, execTimeout, cfg) + if err != nil { + log.Fatalf("Critical error during CronTool initialization: %v", err) + } + + agentLoop.RegisterTool(cronTool) } - agentLoop.RegisterTool(cronTool) - - // Set the onJob handler - cronService.SetOnJob(func(job *cron.CronJob) (string, error) { - result := cronTool.ExecuteJob(context.Background(), job) - return result, nil - }) + // Set onJob handler + if cronTool != nil { + cronService.SetOnJob(func(job *cron.CronJob) (string, error) { + result := cronTool.ExecuteJob(context.Background(), job) + return result, nil + }) + } return cronService } diff --git a/cmd/picoclaw/internal/helpers.go b/cmd/picoclaw/internal/helpers.go index 9655d3c08..f81d7013d 100644 --- a/cmd/picoclaw/internal/helpers.go +++ b/cmd/picoclaw/internal/helpers.go @@ -18,12 +18,21 @@ var ( goVersion string ) +// GetPicoclawHome returns the picoclaw home directory. +// Priority: $PICOCLAW_HOME > ~/.picoclaw +func GetPicoclawHome() string { + if home := os.Getenv("PICOCLAW_HOME"); home != "" { + return home + } + home, _ := os.UserHomeDir() + return filepath.Join(home, ".picoclaw") +} + func GetConfigPath() string { if configPath := os.Getenv("PICOCLAW_CONFIG"); configPath != "" { return configPath } - home, _ := os.UserHomeDir() - return filepath.Join(home, ".picoclaw", "config.json") + return filepath.Join(GetPicoclawHome(), "config.json") } func LoadConfig() (*config.Config, error) { diff --git a/cmd/picoclaw/internal/helpers_test.go b/cmd/picoclaw/internal/helpers_test.go index 47e2f8c07..646be1ba1 100644 --- a/cmd/picoclaw/internal/helpers_test.go +++ b/cmd/picoclaw/internal/helpers_test.go @@ -19,6 +19,27 @@ func TestGetConfigPath(t *testing.T) { assert.Equal(t, want, got) } +func TestGetConfigPath_WithPICOCLAW_HOME(t *testing.T) { + t.Setenv("PICOCLAW_HOME", "/custom/picoclaw") + t.Setenv("HOME", "/tmp/home") + + got := GetConfigPath() + want := filepath.Join("/custom/picoclaw", "config.json") + + assert.Equal(t, want, got) +} + +func TestGetConfigPath_WithPICOCLAW_CONFIG(t *testing.T) { + t.Setenv("PICOCLAW_CONFIG", "/custom/config.json") + t.Setenv("PICOCLAW_HOME", "/custom/picoclaw") + t.Setenv("HOME", "/tmp/home") + + got := GetConfigPath() + want := "/custom/config.json" + + assert.Equal(t, want, got) +} + func TestFormatVersion_NoGitCommit(t *testing.T) { oldVersion, oldGit := version, gitCommit t.Cleanup(func() { version, gitCommit = oldVersion, oldGit }) diff --git a/cmd/picoclaw/internal/skills/install.go b/cmd/picoclaw/internal/skills/install.go index a30f68632..78bc421db 100644 --- a/cmd/picoclaw/internal/skills/install.go +++ b/cmd/picoclaw/internal/skills/install.go @@ -21,8 +21,8 @@ picoclaw skills install --registry clawhub github `, Args: func(cmd *cobra.Command, args []string) error { if registry != "" { - if len(args) != 2 { - return fmt.Errorf("when --registry is set, exactly 2 arguments are required: ") + if len(args) != 1 { + return fmt.Errorf("when --registry is set, exactly 1 argument is required: ") } return nil } @@ -45,7 +45,7 @@ picoclaw skills install --registry clawhub github return err } - return skillsInstallFromRegistry(cfg, args[0], args[1]) + return skillsInstallFromRegistry(cfg, registry, args[0]) } return skillsInstallCmd(installer, args[0]) diff --git a/cmd/picoclaw/internal/skills/install_test.go b/cmd/picoclaw/internal/skills/install_test.go index 97787a986..6b362822d 100644 --- a/cmd/picoclaw/internal/skills/install_test.go +++ b/cmd/picoclaw/internal/skills/install_test.go @@ -26,3 +26,72 @@ func TestNewInstallSubcommand(t *testing.T) { assert.Len(t, cmd.Aliases, 0) } + +func TestInstallCommandArgs(t *testing.T) { + tests := []struct { + name string + args []string + registry string + expectError bool + errorMsg string + }{ + { + name: "no registry, one arg", + args: []string{"sipeed/picoclaw-skills/weather"}, + registry: "", + expectError: false, + }, + { + name: "no registry, no args", + args: []string{}, + registry: "", + expectError: true, + errorMsg: "exactly 1 argument is required: ", + }, + { + name: "no registry, too many args", + args: []string{"arg1", "arg2"}, + registry: "", + expectError: true, + errorMsg: "exactly 1 argument is required: ", + }, + { + name: "with registry, one arg", + args: []string{"weather-skill"}, + registry: "clawhub", + expectError: false, + }, + { + name: "with registry, no args", + args: []string{}, + registry: "clawhub", + expectError: true, + errorMsg: "when --registry is set, exactly 1 argument is required: ", + }, + { + name: "with registry, too many args", + args: []string{"arg1", "arg2"}, + registry: "clawhub", + expectError: true, + errorMsg: "when --registry is set, exactly 1 argument is required: ", + }, + } + + for _, tt := range tests { + t.Run(tt.name, func(t *testing.T) { + cmd := newInstallCommand(nil) + + if tt.registry != "" { + require.NoError(t, cmd.Flags().Set("registry", tt.registry)) + } + + err := cmd.Args(cmd, tt.args) + if tt.expectError { + require.Error(t, err) + assert.Equal(t, tt.errorMsg, err.Error()) + } else { + require.NoError(t, err) + } + }) + } +} diff --git a/config/config.example.json b/config/config.example.json index 3c84cfa9f..2f643d41b 100644 --- a/config/config.example.json +++ b/config/config.example.json @@ -6,7 +6,9 @@ "model_name": "gpt4", "max_tokens": 8192, "temperature": 0.7, - "max_tool_iterations": 20 + "max_tool_iterations": 20, + "summarize_message_threshold": 20, + "summarize_token_percent": 75 } }, "model_list": [ @@ -20,7 +22,8 @@ "model_name": "claude-sonnet-4.6", "model": "anthropic/claude-sonnet-4.6", "api_key": "sk-ant-your-key", - "api_base": "https://api.anthropic.com/v1" + "api_base": "https://api.anthropic.com/v1", + "thinking_level": "high" }, { "model_name": "gemini", @@ -59,6 +62,7 @@ "discord": { "enabled": false, "token": "YOUR_DISCORD_BOT_TOKEN", + "proxy": "", "allow_from": [], "group_trigger": { "mention_only": false @@ -221,27 +225,53 @@ "mistral": { "api_key": "", "api_base": "https://api.mistral.ai/v1" + }, + "avian": { + "api_key": "", + "api_base": "https://api.avian.io/v1" } }, "tools": { + "allow_read_paths": null, + "allow_write_paths": null, "web": { + "enabled": true, "brave": { "enabled": false, "api_key": "YOUR_BRAVE_API_KEY", "max_results": 5 }, + "tavily": { + "enabled": false, + "api_key": "", + "base_url": "", + "max_results": 0 + }, "duckduckgo": { "enabled": true, "max_results": 5 }, "perplexity": { "enabled": false, - "api_key": "pplx-xxx", + "api_key": "", "max_results": 5 }, - "proxy": "" + "searxng": { + "enabled": false, + "base_url": "http://localhost:8888", + "max_results": 5 + }, + "glm_search": { + "enabled": false, + "api_key": "", + "base_url": "https://open.bigmodel.cn/api/paas/v4/web_search", + "search_engine": "search_std", + "max_results": 5 + }, + "fetch_limit_bytes": 10485760 }, "cron": { + "enabled": true, "exec_timeout_minutes": 5 }, "mcp": { @@ -310,19 +340,75 @@ } }, "exec": { - "enable_deny_patterns": false, - "custom_deny_patterns": [] + "enabled": true, + "enable_deny_patterns": true, + "custom_deny_patterns": null, + "custom_allow_patterns": null }, "skills": { + "enabled": true, "registries": { "clawhub": { "enabled": true, "base_url": "https://clawhub.ai", - "search_path": "/api/v1/search", - "skills_path": "/api/v1/skills", - "download_path": "/api/v1/download" + "auth_token": "", + "search_path": "", + "skills_path": "", + "download_path": "", + "timeout": 0, + "max_zip_size": 0, + "max_response_size": 0 } + }, + "max_concurrent_searches": 2, + "search_cache": { + "max_size": 50, + "ttl_seconds": 300 } + }, + "media_cleanup": { + "enabled": true, + "max_age_minutes": 30, + "interval_minutes": 5 + }, + "append_file": { + "enabled": true + }, + "edit_file": { + "enabled": true + }, + "find_skills": { + "enabled": true + }, + "i2c": { + "enabled": false + }, + "install_skill": { + "enabled": true + }, + "list_dir": { + "enabled": true + }, + "message": { + "enabled": true + }, + "read_file": { + "enabled": true + }, + "spawn": { + "enabled": true + }, + "spi": { + "enabled": false + }, + "subagent": { + "enabled": true + }, + "web_fetch": { + "enabled": true + }, + "write_file": { + "enabled": true } }, "heartbeat": { @@ -337,4 +423,4 @@ "host": "127.0.0.1", "port": 18790 } -} \ No newline at end of file +} diff --git a/docs/tools_configuration.md b/docs/tools_configuration.md index 6204fb0c8..e64a3a107 100644 --- a/docs/tools_configuration.md +++ b/docs/tools_configuration.md @@ -180,6 +180,7 @@ The skills tool configures skill discovery and installation via registries like | ---------------------------------- | ------ | -------------------- | ----------------------- | | `registries.clawhub.enabled` | bool | true | Enable ClawHub registry | | `registries.clawhub.base_url` | string | `https://clawhub.ai` | ClawHub base URL | +| `registries.clawhub.auth_token` | string | `""` | Optional Bearer token for higher rate limits | | `registries.clawhub.search_path` | string | `/api/v1/search` | Search API path | | `registries.clawhub.skills_path` | string | `/api/v1/skills` | Skills API path | | `registries.clawhub.download_path` | string | `/api/v1/download` | Download API path | @@ -194,6 +195,7 @@ The skills tool configures skill discovery and installation via registries like "clawhub": { "enabled": true, "base_url": "https://clawhub.ai", + "auth_token": "", "search_path": "/api/v1/search", "skills_path": "/api/v1/skills", "download_path": "/api/v1/download" diff --git a/go.mod b/go.mod index c1172937c..6fa3a900c 100644 --- a/go.mod +++ b/go.mod @@ -11,6 +11,7 @@ require ( github.com/gdamore/tcell/v2 v2.13.8 github.com/google/uuid v1.6.0 github.com/gorilla/websocket v1.5.3 + github.com/h2non/filetype v1.1.3 github.com/larksuite/oapi-sdk-go/v3 v3.5.3 github.com/mdp/qrterminal/v3 v3.2.1 github.com/modelcontextprotocol/go-sdk v1.3.0 @@ -37,8 +38,6 @@ require ( github.com/dustin/go-humanize v1.0.1 // indirect github.com/elliotchance/orderedmap/v3 v3.1.0 // indirect github.com/gdamore/encoding v1.0.1 // indirect - github.com/gdamore/tcell/v2 v2.13.8 // indirect - github.com/h2non/filetype v1.1.3 // indirect github.com/inconshreveable/mousetrap v1.1.0 // indirect github.com/lucasb-eyer/go-colorful v1.3.0 // indirect github.com/mattn/go-colorable v0.1.14 // indirect diff --git a/pkg/agent/context.go b/pkg/agent/context.go index 3aa903b3f..719b0cb6d 100644 --- a/pkg/agent/context.go +++ b/pkg/agent/context.go @@ -42,6 +42,9 @@ type ContextBuilder struct { } func getGlobalConfigDir() string { + if home := os.Getenv("PICOCLAW_HOME"); home != "" { + return home + } home, err := os.UserHomeDir() if err != nil { return "" @@ -602,7 +605,60 @@ func sanitizeHistoryForProvider(history []providers.Message) []providers.Message } } - return sanitized + // Second pass: ensure every assistant message with tool_calls has matching + // tool result messages following it. This is required by strict providers + // like DeepSeek that enforce: "An assistant message with 'tool_calls' must + // be followed by tool messages responding to each 'tool_call_id'." + final := make([]providers.Message, 0, len(sanitized)) + for i := 0; i < len(sanitized); i++ { + msg := sanitized[i] + if msg.Role == "assistant" && len(msg.ToolCalls) > 0 { + // Collect expected tool_call IDs + expected := make(map[string]bool, len(msg.ToolCalls)) + for _, tc := range msg.ToolCalls { + expected[tc.ID] = false + } + + // Check following messages for matching tool results + toolMsgCount := 0 + for j := i + 1; j < len(sanitized); j++ { + if sanitized[j].Role != "tool" { + break + } + toolMsgCount++ + if _, exists := expected[sanitized[j].ToolCallID]; exists { + expected[sanitized[j].ToolCallID] = true + } + } + + // If any tool_call_id is missing, drop this assistant message and its partial tool messages + allFound := true + for toolCallID, found := range expected { + if !found { + allFound = false + logger.DebugCF( + "agent", + "Dropping assistant message with incomplete tool results", + map[string]any{ + "missing_tool_call_id": toolCallID, + "expected_count": len(expected), + "found_count": toolMsgCount, + }, + ) + break + } + } + + if !allFound { + // Skip this assistant message and its tool messages + i += toolMsgCount + continue + } + } + final = append(final, msg) + } + + return final } func (cb *ContextBuilder) AddToolResult( diff --git a/pkg/agent/context_test.go b/pkg/agent/context_test.go index e023c9c30..5756ed911 100644 --- a/pkg/agent/context_test.go +++ b/pkg/agent/context_test.go @@ -207,3 +207,77 @@ func assertRoles(t *testing.T, msgs []providers.Message, expected ...string) { } } } + +// TestSanitizeHistoryForProvider_IncompleteToolResults tests the forward validation +// that ensures assistant messages with tool_calls have ALL matching tool results. +// This fixes the DeepSeek error: "An assistant message with 'tool_calls' must be +// followed by tool messages responding to each 'tool_call_id'." +func TestSanitizeHistoryForProvider_IncompleteToolResults(t *testing.T) { + // Assistant expects tool results for both A and B, but only A is present + history := []providers.Message{ + msg("user", "do two things"), + assistantWithTools("A", "B"), + toolResult("A"), + // toolResult("B") is missing - this would cause DeepSeek to fail + msg("user", "next question"), + msg("assistant", "answer"), + } + + result := sanitizeHistoryForProvider(history) + // The assistant message with incomplete tool results should be dropped, + // along with its partial tool result. The remaining messages are: + // user ("do two things"), user ("next question"), assistant ("answer") + if len(result) != 3 { + t.Fatalf("expected 3 messages, got %d: %+v", len(result), roles(result)) + } + assertRoles(t, result, "user", "user", "assistant") +} + +// TestSanitizeHistoryForProvider_MissingAllToolResults tests the case where +// an assistant message has tool_calls but no tool results follow at all. +func TestSanitizeHistoryForProvider_MissingAllToolResults(t *testing.T) { + history := []providers.Message{ + msg("user", "do something"), + assistantWithTools("A"), + // No tool results at all + msg("user", "hello"), + msg("assistant", "hi"), + } + + result := sanitizeHistoryForProvider(history) + // The assistant message with no tool results should be dropped. + // Remaining: user ("do something"), user ("hello"), assistant ("hi") + if len(result) != 3 { + t.Fatalf("expected 3 messages, got %d: %+v", len(result), roles(result)) + } + assertRoles(t, result, "user", "user", "assistant") +} + +// TestSanitizeHistoryForProvider_PartialToolResultsInMiddle tests that +// incomplete tool results in the middle of a conversation are properly handled. +func TestSanitizeHistoryForProvider_PartialToolResultsInMiddle(t *testing.T) { + history := []providers.Message{ + msg("user", "first"), + assistantWithTools("A"), + toolResult("A"), + msg("assistant", "done"), + msg("user", "second"), + assistantWithTools("B", "C"), + toolResult("B"), + // toolResult("C") is missing + msg("user", "third"), + assistantWithTools("D"), + toolResult("D"), + msg("assistant", "all done"), + } + + result := sanitizeHistoryForProvider(history) + // First round is complete (user, assistant+tools, tool, assistant), + // second round is incomplete and dropped (assistant+tools, partial tool), + // third round is complete (user, assistant+tools, tool, assistant). + // Remaining: user, assistant, tool, assistant, user, user, assistant, tool, assistant + if len(result) != 9 { + t.Fatalf("expected 9 messages, got %d: %+v", len(result), roles(result)) + } + assertRoles(t, result, "user", "assistant", "tool", "assistant", "user", "user", "assistant", "tool", "assistant") +} diff --git a/pkg/agent/instance.go b/pkg/agent/instance.go index ed438059f..97cf0fa05 100644 --- a/pkg/agent/instance.go +++ b/pkg/agent/instance.go @@ -18,22 +18,33 @@ import ( // AgentInstance represents a fully configured agent with its own workspace, // session manager, context builder, and tool registry. type AgentInstance struct { - ID string - Name string - Model string - Fallbacks []string - Workspace string - MaxIterations int - MaxTokens int - Temperature float64 - ContextWindow int - Provider providers.LLMProvider - Sessions *session.SessionManager - ContextBuilder *ContextBuilder - Tools *tools.ToolRegistry - Subagents *config.SubagentsConfig - SkillsFilter []string - Candidates []providers.FallbackCandidate + ID string + Name string + Model string + Fallbacks []string + Workspace string + MaxIterations int + MaxTokens int + Temperature float64 + ThinkingLevel ThinkingLevel + ContextWindow int + SummarizeMessageThreshold int + SummarizeTokenPercent int + Provider providers.LLMProvider + Sessions *session.SessionManager + ContextBuilder *ContextBuilder + Tools *tools.ToolRegistry + Subagents *config.SubagentsConfig + SkillsFilter []string + Candidates []providers.FallbackCandidate + + // Router is non-nil when model routing is configured and the light model + // was successfully resolved. It scores each incoming message and decides + // whether to route to LightCandidates or stay with Candidates. + Router *routing.Router + // LightCandidates holds the resolved provider candidates for the light model. + // Pre-computed at agent creation to avoid repeated model_list lookups at runtime. + LightCandidates []providers.FallbackCandidate } // NewAgentInstance creates an agent instance from config. @@ -57,17 +68,30 @@ func NewAgentInstance( allowWritePaths := compilePatterns(cfg.Tools.AllowWritePaths) toolsRegistry := tools.NewToolRegistry() - toolsRegistry.Register(tools.NewReadFileTool(workspace, readRestrict, allowReadPaths)) - toolsRegistry.Register(tools.NewWriteFileTool(workspace, restrict, allowWritePaths)) - toolsRegistry.Register(tools.NewListDirTool(workspace, readRestrict, allowReadPaths)) - execTool, err := tools.NewExecToolWithConfig(workspace, restrict, cfg) - if err != nil { - log.Fatalf("Critical error: unable to initialize exec tool: %v", err) - } - toolsRegistry.Register(execTool) - toolsRegistry.Register(tools.NewEditFileTool(workspace, restrict, allowWritePaths)) - toolsRegistry.Register(tools.NewAppendFileTool(workspace, restrict, allowWritePaths)) + if cfg.Tools.IsToolEnabled("read_file") { + toolsRegistry.Register(tools.NewReadFileTool(workspace, readRestrict, allowReadPaths)) + } + if cfg.Tools.IsToolEnabled("write_file") { + toolsRegistry.Register(tools.NewWriteFileTool(workspace, restrict, allowWritePaths)) + } + if cfg.Tools.IsToolEnabled("list_dir") { + toolsRegistry.Register(tools.NewListDirTool(workspace, readRestrict, allowReadPaths)) + } + if cfg.Tools.IsToolEnabled("exec") { + execTool, err := tools.NewExecToolWithConfig(workspace, restrict, cfg) + if err != nil { + log.Fatalf("Critical error: unable to initialize exec tool: %v", err) + } + toolsRegistry.Register(execTool) + } + + if cfg.Tools.IsToolEnabled("edit_file") { + toolsRegistry.Register(tools.NewEditFileTool(workspace, restrict, allowWritePaths)) + } + if cfg.Tools.IsToolEnabled("append_file") { + toolsRegistry.Register(tools.NewAppendFileTool(workspace, restrict, allowWritePaths)) + } sessionsDir := filepath.Join(workspace, "sessions") sessionsManager := session.NewSessionManager(sessionsDir) @@ -101,6 +125,22 @@ func NewAgentInstance( temperature = *defaults.Temperature } + var thinkingLevelStr string + if mc, err := cfg.GetModelConfig(model); err == nil { + thinkingLevelStr = mc.ThinkingLevel + } + thinkingLevel := parseThinkingLevel(thinkingLevelStr) + + summarizeMessageThreshold := defaults.SummarizeMessageThreshold + if summarizeMessageThreshold == 0 { + summarizeMessageThreshold = 20 + } + + summarizeTokenPercent := defaults.SummarizeTokenPercent + if summarizeTokenPercent == 0 { + summarizeTokenPercent = 75 + } + // Resolve fallback candidates modelCfg := providers.ModelConfig{ Primary: model, @@ -148,23 +188,47 @@ func NewAgentInstance( candidates := providers.ResolveCandidatesWithLookup(modelCfg, defaults.Provider, resolveFromModelList) + // Model routing setup: pre-resolve light model candidates at creation time + // to avoid repeated model_list lookups on every incoming message. + var router *routing.Router + var lightCandidates []providers.FallbackCandidate + if rc := defaults.Routing; rc != nil && rc.Enabled && rc.LightModel != "" { + lightModelCfg := providers.ModelConfig{Primary: rc.LightModel} + resolved := providers.ResolveCandidatesWithLookup(lightModelCfg, defaults.Provider, resolveFromModelList) + if len(resolved) > 0 { + router = routing.New(routing.RouterConfig{ + LightModel: rc.LightModel, + Threshold: rc.Threshold, + }) + lightCandidates = resolved + } else { + log.Printf("routing: light_model %q not found in model_list — routing disabled for agent %q", + rc.LightModel, agentID) + } + } + return &AgentInstance{ - ID: agentID, - Name: agentName, - Model: model, - Fallbacks: fallbacks, - Workspace: workspace, - MaxIterations: maxIter, - MaxTokens: maxTokens, - Temperature: temperature, - ContextWindow: maxTokens, - Provider: provider, - Sessions: sessionsManager, - ContextBuilder: contextBuilder, - Tools: toolsRegistry, - Subagents: subagents, - SkillsFilter: skillsFilter, - Candidates: candidates, + ID: agentID, + Name: agentName, + Model: model, + Fallbacks: fallbacks, + Workspace: workspace, + MaxIterations: maxIter, + MaxTokens: maxTokens, + Temperature: temperature, + ThinkingLevel: thinkingLevel, + ContextWindow: maxTokens, + SummarizeMessageThreshold: summarizeMessageThreshold, + SummarizeTokenPercent: summarizeTokenPercent, + Provider: provider, + Sessions: sessionsManager, + ContextBuilder: contextBuilder, + Tools: toolsRegistry, + Subagents: subagents, + SkillsFilter: skillsFilter, + Candidates: candidates, + Router: router, + LightCandidates: lightCandidates, } } @@ -173,12 +237,13 @@ func resolveAgentWorkspace(agentCfg *config.AgentConfig, defaults *config.AgentD if agentCfg != nil && strings.TrimSpace(agentCfg.Workspace) != "" { return expandHome(strings.TrimSpace(agentCfg.Workspace)) } + // Use the configured default workspace (respects PICOCLAW_HOME) if agentCfg == nil || agentCfg.Default || agentCfg.ID == "" || routing.NormalizeAgentID(agentCfg.ID) == "main" { return expandHome(defaults.Workspace) } - home, _ := os.UserHomeDir() + // For named agents without explicit workspace, use default workspace with agent ID suffix id := routing.NormalizeAgentID(agentCfg.ID) - return filepath.Join(home, ".picoclaw", "workspace-"+id) + return filepath.Join(expandHome(defaults.Workspace), "..", "workspace-"+id) } // resolveAgentModel resolves the primary model for an agent. diff --git a/pkg/agent/loop.go b/pkg/agent/loop.go index 62b2d48c3..9352150df 100644 --- a/pkg/agent/loop.go +++ b/pkg/agent/loop.go @@ -12,6 +12,7 @@ import ( "errors" "fmt" "path/filepath" + "regexp" "strings" "sync" "sync/atomic" @@ -20,6 +21,7 @@ import ( "github.com/sipeed/picoclaw/pkg/bus" "github.com/sipeed/picoclaw/pkg/channels" + "github.com/sipeed/picoclaw/pkg/commands" "github.com/sipeed/picoclaw/pkg/config" "github.com/sipeed/picoclaw/pkg/constants" "github.com/sipeed/picoclaw/pkg/logger" @@ -31,6 +33,7 @@ import ( "github.com/sipeed/picoclaw/pkg/state" "github.com/sipeed/picoclaw/pkg/tools" "github.com/sipeed/picoclaw/pkg/utils" + "github.com/sipeed/picoclaw/pkg/voice" ) type AgentLoop struct { @@ -43,6 +46,8 @@ type AgentLoop struct { fallback *providers.FallbackChain channelManager *channels.Manager mediaStore media.MediaStore + transcriber voice.Transcriber + cmdRegistry *commands.Registry } // processOptions configures how a message is processed @@ -58,7 +63,15 @@ type processOptions struct { NoHistory bool // If true, don't load session history (for heartbeat) } -const defaultResponse = "I've completed processing but have no response to give. Increase `max_tool_iterations` in config.json." +const ( + defaultResponse = "I've completed processing but have no response to give. Increase `max_tool_iterations` in config.json." + sessionKeyAgentPrefix = "agent:" + metadataKeyAccountID = "account_id" + metadataKeyGuildID = "guild_id" + metadataKeyTeamID = "team_id" + metadataKeyParentPeerKind = "parent_peer_kind" + metadataKeyParentPeerID = "parent_peer_id" +) func NewAgentLoop( cfg *config.Config, @@ -81,14 +94,17 @@ func NewAgentLoop( stateManager = state.NewManager(defaultAgent.Workspace) } - return &AgentLoop{ + al := &AgentLoop{ bus: msgBus, cfg: cfg, registry: registry, state: stateManager, summarizing: sync.Map{}, fallback: fallbackChain, + cmdRegistry: commands.NewRegistry(commands.BuiltinDefinitions()), } + + return al } // registerSharedTools registers tools that are shared across all agents (web, message, spawn). @@ -105,71 +121,117 @@ func registerSharedTools( } // Web tools - searchTool, err := tools.NewWebSearchTool(tools.WebSearchToolOptions{ - BraveAPIKey: cfg.Tools.Web.Brave.APIKey, - BraveMaxResults: cfg.Tools.Web.Brave.MaxResults, - BraveEnabled: cfg.Tools.Web.Brave.Enabled, - TavilyAPIKey: cfg.Tools.Web.Tavily.APIKey, - TavilyBaseURL: cfg.Tools.Web.Tavily.BaseURL, - TavilyMaxResults: cfg.Tools.Web.Tavily.MaxResults, - TavilyEnabled: cfg.Tools.Web.Tavily.Enabled, - DuckDuckGoMaxResults: cfg.Tools.Web.DuckDuckGo.MaxResults, - DuckDuckGoEnabled: cfg.Tools.Web.DuckDuckGo.Enabled, - PerplexityAPIKey: cfg.Tools.Web.Perplexity.APIKey, - PerplexityMaxResults: cfg.Tools.Web.Perplexity.MaxResults, - PerplexityEnabled: cfg.Tools.Web.Perplexity.Enabled, - Proxy: cfg.Tools.Web.Proxy, - }) - if err != nil { - logger.ErrorCF("agent", "Failed to create web search tool", map[string]any{"error": err.Error()}) - } else if searchTool != nil { - agent.Tools.Register(searchTool) + if cfg.Tools.IsToolEnabled("web") { + searchTool, err := tools.NewWebSearchTool(tools.WebSearchToolOptions{ + BraveAPIKey: cfg.Tools.Web.Brave.APIKey, + BraveMaxResults: cfg.Tools.Web.Brave.MaxResults, + BraveEnabled: cfg.Tools.Web.Brave.Enabled, + TavilyAPIKey: cfg.Tools.Web.Tavily.APIKey, + TavilyBaseURL: cfg.Tools.Web.Tavily.BaseURL, + TavilyMaxResults: cfg.Tools.Web.Tavily.MaxResults, + TavilyEnabled: cfg.Tools.Web.Tavily.Enabled, + DuckDuckGoMaxResults: cfg.Tools.Web.DuckDuckGo.MaxResults, + DuckDuckGoEnabled: cfg.Tools.Web.DuckDuckGo.Enabled, + PerplexityAPIKey: cfg.Tools.Web.Perplexity.APIKey, + PerplexityMaxResults: cfg.Tools.Web.Perplexity.MaxResults, + PerplexityEnabled: cfg.Tools.Web.Perplexity.Enabled, + SearXNGBaseURL: cfg.Tools.Web.SearXNG.BaseURL, + SearXNGMaxResults: cfg.Tools.Web.SearXNG.MaxResults, + SearXNGEnabled: cfg.Tools.Web.SearXNG.Enabled, + GLMSearchAPIKey: cfg.Tools.Web.GLMSearch.APIKey, + GLMSearchBaseURL: cfg.Tools.Web.GLMSearch.BaseURL, + GLMSearchEngine: cfg.Tools.Web.GLMSearch.SearchEngine, + GLMSearchMaxResults: cfg.Tools.Web.GLMSearch.MaxResults, + GLMSearchEnabled: cfg.Tools.Web.GLMSearch.Enabled, + Proxy: cfg.Tools.Web.Proxy, + }) + if err != nil { + logger.ErrorCF("agent", "Failed to create web search tool", map[string]any{"error": err.Error()}) + } else if searchTool != nil { + agent.Tools.Register(searchTool) + } } - fetchTool, err := tools.NewWebFetchToolWithProxy(50000, cfg.Tools.Web.Proxy, cfg.Tools.Web.FetchLimitBytes) - if err != nil { - logger.ErrorCF("agent", "Failed to create web fetch tool", map[string]any{"error": err.Error()}) - } else { - agent.Tools.Register(fetchTool) + if cfg.Tools.IsToolEnabled("web_fetch") { + fetchTool, err := tools.NewWebFetchToolWithProxy(50000, cfg.Tools.Web.Proxy, cfg.Tools.Web.FetchLimitBytes) + if err != nil { + logger.ErrorCF("agent", "Failed to create web fetch tool", map[string]any{"error": err.Error()}) + } else { + agent.Tools.Register(fetchTool) + } } // Hardware tools (I2C, SPI) - Linux only, returns error on other platforms - agent.Tools.Register(tools.NewI2CTool()) - agent.Tools.Register(tools.NewSPITool()) + if cfg.Tools.IsToolEnabled("i2c") { + agent.Tools.Register(tools.NewI2CTool()) + } + if cfg.Tools.IsToolEnabled("spi") { + agent.Tools.Register(tools.NewSPITool()) + } // Message tool - messageTool := tools.NewMessageTool() - messageTool.SetSendCallback(func(channel, chatID, content string) error { - pubCtx, pubCancel := context.WithTimeout(context.Background(), 5*time.Second) - defer pubCancel() - return msgBus.PublishOutbound(pubCtx, bus.OutboundMessage{ - Channel: channel, - ChatID: chatID, - Content: content, + if cfg.Tools.IsToolEnabled("message") { + messageTool := tools.NewMessageTool() + messageTool.SetSendCallback(func(channel, chatID, content string) error { + pubCtx, pubCancel := context.WithTimeout(context.Background(), 5*time.Second) + defer pubCancel() + return msgBus.PublishOutbound(pubCtx, bus.OutboundMessage{ + Channel: channel, + ChatID: chatID, + Content: content, + }) }) - }) - agent.Tools.Register(messageTool) + agent.Tools.Register(messageTool) + } + + // Send file tool (outbound media via MediaStore — store injected later by SetMediaStore) + if cfg.Tools.IsToolEnabled("send_file") { + sendFileTool := tools.NewSendFileTool( + agent.Workspace, + cfg.Agents.Defaults.RestrictToWorkspace, + cfg.Agents.Defaults.GetMaxMediaSize(), + nil, + ) + agent.Tools.Register(sendFileTool) + } // Skill discovery and installation tools - registryMgr := skills.NewRegistryManagerFromConfig(skills.RegistryConfig{ - MaxConcurrentSearches: cfg.Tools.Skills.MaxConcurrentSearches, - ClawHub: skills.ClawHubConfig(cfg.Tools.Skills.Registries.ClawHub), - }) - searchCache := skills.NewSearchCache( - cfg.Tools.Skills.SearchCache.MaxSize, - time.Duration(cfg.Tools.Skills.SearchCache.TTLSeconds)*time.Second, - ) - agent.Tools.Register(tools.NewFindSkillsTool(registryMgr, searchCache)) - agent.Tools.Register(tools.NewInstallSkillTool(registryMgr, agent.Workspace)) + skills_enabled := cfg.Tools.IsToolEnabled("skills") + find_skills_enable := cfg.Tools.IsToolEnabled("find_skills") + install_skills_enable := cfg.Tools.IsToolEnabled("install_skill") + if skills_enabled && (find_skills_enable || install_skills_enable) { + registryMgr := skills.NewRegistryManagerFromConfig(skills.RegistryConfig{ + MaxConcurrentSearches: cfg.Tools.Skills.MaxConcurrentSearches, + ClawHub: skills.ClawHubConfig(cfg.Tools.Skills.Registries.ClawHub), + }) + + if find_skills_enable { + searchCache := skills.NewSearchCache( + cfg.Tools.Skills.SearchCache.MaxSize, + time.Duration(cfg.Tools.Skills.SearchCache.TTLSeconds)*time.Second, + ) + agent.Tools.Register(tools.NewFindSkillsTool(registryMgr, searchCache)) + } + + if install_skills_enable { + agent.Tools.Register(tools.NewInstallSkillTool(registryMgr, agent.Workspace)) + } + } // Spawn tool with allowlist checker - subagentManager := tools.NewSubagentManager(provider, agent.Model, agent.Workspace, msgBus) - subagentManager.SetLLMOptions(agent.MaxTokens, agent.Temperature) - spawnTool := tools.NewSpawnTool(subagentManager) - currentAgentID := agentID - spawnTool.SetAllowlistChecker(func(targetAgentID string) bool { - return registry.CanSpawnSubagent(currentAgentID, targetAgentID) - }) - agent.Tools.Register(spawnTool) + if cfg.Tools.IsToolEnabled("spawn") { + if cfg.Tools.IsToolEnabled("subagent") { + subagentManager := tools.NewSubagentManager(provider, agent.Model, agent.Workspace, msgBus) + subagentManager.SetLLMOptions(agent.MaxTokens, agent.Temperature) + spawnTool := tools.NewSpawnTool(subagentManager) + currentAgentID := agentID + spawnTool.SetAllowlistChecker(func(targetAgentID string) bool { + return registry.CanSpawnSubagent(currentAgentID, targetAgentID) + }) + agent.Tools.Register(spawnTool) + } else { + logger.WarnCF("agent", "spawn tool requires subagent to be enabled", nil) + } + } } } @@ -177,7 +239,7 @@ func (al *AgentLoop) Run(ctx context.Context) error { al.running.Store(true) // Initialize MCP servers for all agents - if al.cfg.Tools.MCP.Enabled { + if al.cfg.Tools.IsToolEnabled("mcp") { mcpManager := mcp.NewManager() // Ensure MCP connections are cleaned up on exit, regardless of initialization success // This fixes resource leak when LoadFromMCPConfig partially succeeds then fails @@ -219,6 +281,7 @@ func (al *AgentLoop) Run(ctx context.Context) error { if !ok { continue } + mcpTool := tools.NewMCPTool(mcpManager, serverName, tool) agent.Tools.Register(mcpTool) totalRegistrations++ @@ -332,6 +395,71 @@ func (al *AgentLoop) SetChannelManager(cm *channels.Manager) { // SetMediaStore injects a MediaStore for media lifecycle management. func (al *AgentLoop) SetMediaStore(s media.MediaStore) { al.mediaStore = s + + // Propagate store to send_file tools in all agents. + al.registry.ForEachTool("send_file", func(t tools.Tool) { + if sf, ok := t.(*tools.SendFileTool); ok { + sf.SetMediaStore(s) + } + }) +} + +// SetTranscriber injects a voice transcriber for agent-level audio transcription. +func (al *AgentLoop) SetTranscriber(t voice.Transcriber) { + al.transcriber = t +} + +var audioAnnotationRe = regexp.MustCompile(`\[(voice|audio)(?::[^\]]*)?\]`) + +// transcribeAudioInMessage resolves audio media refs, transcribes them, and +// replaces audio annotations in msg.Content with the transcribed text. +func (al *AgentLoop) transcribeAudioInMessage(ctx context.Context, msg bus.InboundMessage) bus.InboundMessage { + if al.transcriber == nil || al.mediaStore == nil || len(msg.Media) == 0 { + return msg + } + + // Transcribe each audio media ref in order. + var transcriptions []string + for _, ref := range msg.Media { + path, meta, err := al.mediaStore.ResolveWithMeta(ref) + if err != nil { + logger.WarnCF("voice", "Failed to resolve media ref", map[string]any{"ref": ref, "error": err}) + continue + } + if !utils.IsAudioFile(meta.Filename, meta.ContentType) { + continue + } + result, err := al.transcriber.Transcribe(ctx, path) + if err != nil { + logger.WarnCF("voice", "Transcription failed", map[string]any{"ref": ref, "error": err}) + transcriptions = append(transcriptions, "") + continue + } + transcriptions = append(transcriptions, result.Text) + } + + if len(transcriptions) == 0 { + return msg + } + + // Replace audio annotations sequentially with transcriptions. + idx := 0 + newContent := audioAnnotationRe.ReplaceAllStringFunc(msg.Content, func(match string) string { + if idx >= len(transcriptions) { + return match + } + text := transcriptions[idx] + idx++ + return "[voice: " + text + "]" + }) + + // Append any remaining transcriptions not matched by an annotation. + for ; idx < len(transcriptions); idx++ { + newContent += "\n[voice: " + transcriptions[idx] + "]" + } + + msg.Content = newContent + return msg } // inferMediaType determines the media type ("image", "audio", "video", "file") @@ -445,52 +573,46 @@ func (al *AgentLoop) processMessage(ctx context.Context, msg bus.InboundMessage) }, ) + msg = al.transcribeAudioInMessage(ctx, msg) + // Route system messages to processSystemMessage if msg.Channel == "system" { return al.processSystemMessage(ctx, msg) } - // Check for commands - if response, handled := al.handleCommand(ctx, msg); handled { + route, agent, routeErr := al.resolveMessageRoute(msg) + + // Commands are checked before requiring a successful route. + // Global commands (/help, /show, /switch) work even when routing fails; + // context-dependent commands check their own Runtime fields and report + // "unavailable" when the required capability is nil. + if response, handled := al.handleCommand(ctx, msg, agent); handled { return response, nil } - // Route to determine agent and session key - route := al.registry.ResolveRoute(routing.RouteInput{ - Channel: msg.Channel, - AccountID: msg.Metadata["account_id"], - Peer: extractPeer(msg), - ParentPeer: extractParentPeer(msg), - GuildID: msg.Metadata["guild_id"], - TeamID: msg.Metadata["team_id"], - }) - - agent, ok := al.registry.GetAgent(route.AgentID) - if !ok { - agent = al.registry.GetDefaultAgent() - } - if agent == nil { - return "", fmt.Errorf("no agent available for route (agent_id=%s)", route.AgentID) + if routeErr != nil { + return "", routeErr } // Reset message-tool state for this round so we don't skip publishing due to a previous round. if tool, ok := agent.Tools.Get("message"); ok { - if mt, ok := tool.(tools.ContextualTool); ok { - mt.SetContext(msg.Channel, msg.ChatID) + if resetter, ok := tool.(interface{ ResetSentInRound() }); ok { + resetter.ResetSentInRound() } } - // Use routed session key, but honor pre-set agent-scoped keys (for ProcessDirect/cron) - sessionKey := route.SessionKey - if msg.SessionKey != "" && strings.HasPrefix(msg.SessionKey, "agent:") { - sessionKey = msg.SessionKey - } + // Resolve session key from route, while preserving explicit agent-scoped keys. + scopeKey := resolveScopeKey(route, msg.SessionKey) + sessionKey := scopeKey logger.InfoCF("agent", "Routed message", map[string]any{ - "agent_id": agent.ID, - "session_key": sessionKey, - "matched_by": route.MatchedBy, + "agent_id": agent.ID, + "scope_key": scopeKey, + "session_key": sessionKey, + "matched_by": route.MatchedBy, + "route_agent": route.AgentID, + "route_channel": route.Channel, }) return al.runAgentLoop(ctx, agent, processOptions{ @@ -505,6 +627,34 @@ func (al *AgentLoop) processMessage(ctx context.Context, msg bus.InboundMessage) }) } +func (al *AgentLoop) resolveMessageRoute(msg bus.InboundMessage) (routing.ResolvedRoute, *AgentInstance, error) { + route := al.registry.ResolveRoute(routing.RouteInput{ + Channel: msg.Channel, + AccountID: inboundMetadata(msg, metadataKeyAccountID), + Peer: extractPeer(msg), + ParentPeer: extractParentPeer(msg), + GuildID: inboundMetadata(msg, metadataKeyGuildID), + TeamID: inboundMetadata(msg, metadataKeyTeamID), + }) + + agent, ok := al.registry.GetAgent(route.AgentID) + if !ok { + agent = al.registry.GetDefaultAgent() + } + if agent == nil { + return routing.ResolvedRoute{}, nil, fmt.Errorf("no agent available for route (agent_id=%s)", route.AgentID) + } + + return route, agent, nil +} + +func resolveScopeKey(route routing.ResolvedRoute, msgSessionKey string) string { + if msgSessionKey != "" && strings.HasPrefix(msgSessionKey, sessionKeyAgentPrefix) { + return msgSessionKey + } + return route.SessionKey +} + func (al *AgentLoop) processSystemMessage( ctx context.Context, msg bus.InboundMessage, @@ -591,10 +741,7 @@ func (al *AgentLoop) runAgentLoop( } } - // 1. Update tool contexts - al.updateToolContexts(agent, opts.Channel, opts.ChatID) - - // 2. Build messages (skip history for heartbeat) + // 1. Build messages (skip history for heartbeat) var history []providers.Message var summary string if !opts.NoHistory { @@ -614,7 +761,7 @@ func (al *AgentLoop) runAgentLoop( maxMediaSize := al.cfg.Agents.Defaults.GetMaxMediaSize() messages = resolveMediaRefs(messages, al.mediaStore, maxMediaSize) - // 3. Save user message to session + // 2. Save user message to session agent.Sessions.AddMessage(opts.SessionKey, "user", opts.UserMessage) // 4. Run LLM iteration loop @@ -626,7 +773,7 @@ func (al *AgentLoop) runAgentLoop( // If last tool had ForUser content and we already sent it, we might not need to send final response // This is controlled by the tool's Silent flag and ForUser content - // 5. Handle empty response + // 4. Handle empty response if finalContent == "" { // Deduplicate tools for TLDR seen := make(map[string]bool) @@ -640,16 +787,16 @@ func (al *AgentLoop) runAgentLoop( finalContent = generateTLDR(opts.UserMessage, uniqueTools, iteration, al.cfg.Agents.Defaults.TLDRIncludeMessage) } - // 6. Save final assistant message to session + // 5. Save final assistant message to session agent.Sessions.AddMessage(opts.SessionKey, "assistant", finalContent) agent.Sessions.Save(opts.SessionKey) - // 7. Optional: summarization + // 6. Optional: summarization if opts.EnableSummary { al.maybeSummarize(agent, opts.SessionKey, opts.Channel, opts.ChatID) } - // 8. Optional: send response via bus + // 7. Optional: send response via bus if opts.SendResponse { al.bus.PublishOutbound(ctx, bus.OutboundMessage{ Channel: opts.Channel, @@ -658,7 +805,7 @@ func (al *AgentLoop) runAgentLoop( }) } - // 9. Log response + // 8. Log response responsePreview := utils.Truncate(finalContent, 120) logger.InfoCF("agent", fmt.Sprintf("Response: %s", responsePreview), map[string]any{ @@ -738,6 +885,12 @@ func (al *AgentLoop) runLLMIteration( var finalContent string executedTools := []string{} + // Determine effective model tier for this conversation turn. + // selectCandidates evaluates routing once and the decision is sticky for + // all tool-follow-up iterations within the same turn so that a multi-step + // tool chain doesn't switch models mid-way through. + activeCandidates, activeModel := al.selectCandidates(agent, opts.UserMessage, messages) + for iteration < agent.MaxIterations { iteration++ @@ -756,7 +909,7 @@ func (al *AgentLoop) runLLMIteration( map[string]any{ "agent_id": agent.ID, "iteration": iteration, - "model": agent.Model, + "model": activeModel, "messages_count": len(messages), "tools_count": len(providerToolDefs), "max_tokens": agent.MaxTokens, @@ -772,27 +925,33 @@ func (al *AgentLoop) runLLMIteration( "tools_json": formatToolsForLog(providerToolDefs), }) - // Call LLM with fallback chain if candidates are configured. + // Call LLM with fallback chain if multiple candidates are configured. var response *providers.LLMResponse var err error + llmOpts := map[string]any{ + "max_tokens": agent.MaxTokens, + "temperature": agent.Temperature, + "prompt_cache_key": agent.ID, + } + // parseThinkingLevel guarantees ThinkingOff for empty/unknown values, + // so checking != ThinkingOff is sufficient. + if agent.ThinkingLevel != ThinkingOff { + if tc, ok := agent.Provider.(providers.ThinkingCapable); ok && tc.SupportsThinking() { + llmOpts["thinking_level"] = string(agent.ThinkingLevel) + } else { + logger.WarnCF("agent", "thinking_level is set but current provider does not support it, ignoring", + map[string]any{"agent_id": agent.ID, "thinking_level": string(agent.ThinkingLevel)}) + } + } + callLLM := func() (*providers.LLMResponse, error) { - if len(agent.Candidates) > 1 && al.fallback != nil { + if len(activeCandidates) > 1 && al.fallback != nil { fbResult, fbErr := al.fallback.Execute( ctx, - agent.Candidates, + activeCandidates, func(ctx context.Context, provider, model string) (*providers.LLMResponse, error) { - return agent.Provider.Chat( - ctx, - messages, - providerToolDefs, - model, - map[string]any{ - "max_tokens": agent.MaxTokens, - "temperature": agent.Temperature, - "prompt_cache_key": agent.ID, - }, - ) + return agent.Provider.Chat(ctx, messages, providerToolDefs, model, llmOpts) }, ) if fbErr != nil { @@ -808,11 +967,7 @@ func (al *AgentLoop) runLLMIteration( } return fbResult.Response, nil } - return agent.Provider.Chat(ctx, messages, providerToolDefs, agent.Model, map[string]any{ - "max_tokens": agent.MaxTokens, - "temperature": agent.Temperature, - "prompt_cache_key": agent.ID, - }) + return agent.Provider.Chat(ctx, messages, providerToolDefs, activeModel, llmOpts) } // Retry loop for context/token errors @@ -891,7 +1046,7 @@ func (al *AgentLoop) runLLMIteration( "iteration": iteration, "error": err.Error(), }) - return "", iteration, fmt.Errorf("LLM call failed after retries: %w", err) + return "", iteration, nil, fmt.Errorf("LLM call failed after retries: %w", err) } go al.handleReasoning( @@ -977,62 +1132,76 @@ func (al *AgentLoop) runLLMIteration( // Save assistant message with tool calls to session agent.Sessions.AddFullMessage(opts.SessionKey, assistantMsg) - // Execute tool calls - for _, tc := range normalizedToolCalls { - argsJSON, _ := json.Marshal(tc.Arguments) - argsPreview := utils.Truncate(string(argsJSON), 200) - logger.InfoCF("agent", fmt.Sprintf("Tool call: %s(%s)", tc.Name, argsPreview), - map[string]any{ - "agent_id": agent.ID, - "tool": tc.Name, - "iteration": iteration, - }) + // Execute tool calls in parallel + type indexedAgentResult struct { + result *tools.ToolResult + tc providers.ToolCall + } - // Create async callback for tools that implement AsyncTool - // NOTE: Following openclaw's design, async tools do NOT send results directly to users. - // Instead, they notify the agent via PublishInbound, and the agent decides - // whether to forward the result to the user (in processSystemMessage). - asyncCallback := func(callbackCtx context.Context, result *tools.ToolResult) { - // Log the async completion but don't send directly to user - // The agent will handle user notification via processSystemMessage - if !result.Silent && result.ForUser != "" { - logger.InfoCF("agent", "Async tool completed, agent will handle notification", - map[string]any{ - "tool": tc.Name, - "content_len": len(result.ForUser), - }) + agentResults := make([]indexedAgentResult, len(normalizedToolCalls)) + var wg sync.WaitGroup + + for i, tc := range normalizedToolCalls { + agentResults[i].tc = tc + + wg.Add(1) + go func(idx int, tc providers.ToolCall) { + defer wg.Done() + + argsJSON, _ := json.Marshal(tc.Arguments) + argsPreview := utils.Truncate(string(argsJSON), 200) + logger.InfoCF("agent", fmt.Sprintf("Tool call: %s(%s)", tc.Name, argsPreview), + map[string]any{ + "agent_id": agent.ID, + "tool": tc.Name, + "iteration": iteration, + }) + + // Create async callback for tools that implement AsyncExecutor + asyncCallback := func(callbackCtx context.Context, result *tools.ToolResult) { + if !result.Silent && result.ForUser != "" { + logger.InfoCF("agent", "Async tool completed, agent will handle notification", + map[string]any{ + "tool": tc.Name, + "content_len": len(result.ForUser), + }) + } } - } - toolResult := agent.Tools.ExecuteWithContext( - ctx, - tc.Name, - tc.Arguments, - opts.Channel, - opts.ChatID, - asyncCallback, - ) + toolResult := agent.Tools.ExecuteWithContext( + ctx, + tc.Name, + tc.Arguments, + opts.Channel, + opts.ChatID, + asyncCallback, + ) + agentResults[idx].result = toolResult + }(i, tc) + } + wg.Wait() + // Process results in original order (send to user, save to session) + for _, r := range agentResults { // Send ForUser content to user immediately if not Silent - if !toolResult.Silent && toolResult.ForUser != "" && opts.SendResponse { + if !r.result.Silent && r.result.ForUser != "" && opts.SendResponse { al.bus.PublishOutbound(ctx, bus.OutboundMessage{ Channel: opts.Channel, ChatID: opts.ChatID, - Content: toolResult.ForUser, + Content: r.result.ForUser, }) logger.DebugCF("agent", "Sent tool result to user", map[string]any{ - "tool": tc.Name, - "content_len": len(toolResult.ForUser), + "tool": r.tc.Name, + "content_len": len(r.result.ForUser), }) } // If tool returned media refs, publish them as outbound media - if len(toolResult.Media) > 0 && opts.SendResponse { - parts := make([]bus.MediaPart, 0, len(toolResult.Media)) - for _, ref := range toolResult.Media { + if len(r.result.Media) > 0 { + parts := make([]bus.MediaPart, 0, len(r.result.Media)) + for _, ref := range r.result.Media { part := bus.MediaPart{Ref: ref} - // Populate metadata from MediaStore when available if al.mediaStore != nil { if _, meta, err := al.mediaStore.ResolveWithMeta(ref); err == nil { part.Filename = meta.Filename @@ -1050,15 +1219,15 @@ func (al *AgentLoop) runLLMIteration( } // Determine content for LLM based on tool result - contentForLLM := toolResult.ForLLM - if contentForLLM == "" && toolResult.Err != nil { - contentForLLM = toolResult.Err.Error() + contentForLLM := r.result.ForLLM + if contentForLLM == "" && r.result.Err != nil { + contentForLLM = r.result.Err.Error() } toolResultMsg := providers.Message{ Role: "tool", Content: contentForLLM, - ToolCallID: tc.ID, + ToolCallID: r.tc.ID, } messages = append(messages, toolResultMsg) @@ -1067,36 +1236,54 @@ func (al *AgentLoop) runLLMIteration( } } - return finalContent, iteration, nil + return finalContent, iteration, executedTools, nil } -// updateToolContexts updates the context for tools that need channel/chatID info. -func (al *AgentLoop) updateToolContexts(agent *AgentInstance, channel, chatID string) { - // Use ContextualTool interface instead of type assertions - if tool, ok := agent.Tools.Get("message"); ok { - if mt, ok := tool.(tools.ContextualTool); ok { - mt.SetContext(channel, chatID) - } +// selectCandidates returns the model candidates and resolved model name to use +// for a conversation turn. When model routing is configured and the incoming +// message scores below the complexity threshold, it returns the light model +// candidates instead of the primary ones. +// +// The returned (candidates, model) pair is used for all LLM calls within one +// turn — tool follow-up iterations use the same tier as the initial call so +// that a multi-step tool chain doesn't switch models mid-way. +func (al *AgentLoop) selectCandidates( + agent *AgentInstance, + userMsg string, + history []providers.Message, +) (candidates []providers.FallbackCandidate, model string) { + if agent.Router == nil || len(agent.LightCandidates) == 0 { + return agent.Candidates, agent.Model } - if tool, ok := agent.Tools.Get("spawn"); ok { - if st, ok := tool.(tools.ContextualTool); ok { - st.SetContext(channel, chatID) - } - } - if tool, ok := agent.Tools.Get("subagent"); ok { - if st, ok := tool.(tools.ContextualTool); ok { - st.SetContext(channel, chatID) - } + + _, usedLight, score := agent.Router.SelectModel(userMsg, history, agent.Model) + if !usedLight { + logger.DebugCF("agent", "Model routing: primary model selected", + map[string]any{ + "agent_id": agent.ID, + "score": score, + "threshold": agent.Router.Threshold(), + }) + return agent.Candidates, agent.Model } + + logger.InfoCF("agent", "Model routing: light model selected", + map[string]any{ + "agent_id": agent.ID, + "light_model": agent.Router.LightModel(), + "score": score, + "threshold": agent.Router.Threshold(), + }) + return agent.LightCandidates, agent.Router.LightModel() } // maybeSummarize triggers summarization if the session history exceeds thresholds. func (al *AgentLoop) maybeSummarize(agent *AgentInstance, sessionKey, channel, chatID string) { newHistory := agent.Sessions.GetHistory(sessionKey) tokenEstimate := al.estimateTokens(newHistory) - threshold := agent.ContextWindow * 75 / 100 + threshold := agent.ContextWindow * agent.SummarizeTokenPercent / 100 - if len(newHistory) > 20 || tokenEstimate > threshold { + if len(newHistory) > agent.SummarizeMessageThreshold || tokenEstimate > threshold { summarizeKey := agent.ID + ":" + sessionKey if _, loading := al.summarizing.LoadOrStore(summarizeKey, true); !loading { go func() { @@ -1381,94 +1568,87 @@ func (al *AgentLoop) estimateTokens(messages []providers.Message) int { return totalChars * 2 / 5 } -func (al *AgentLoop) handleCommand(ctx context.Context, msg bus.InboundMessage) (string, bool) { - content := strings.TrimSpace(msg.Content) - if !strings.HasPrefix(content, "/") { +func (al *AgentLoop) handleCommand( + ctx context.Context, + msg bus.InboundMessage, + agent *AgentInstance, +) (string, bool) { + if !commands.HasCommandPrefix(msg.Content) { return "", false } - parts := strings.Fields(content) - if len(parts) == 0 { + if al.cmdRegistry == nil { return "", false } - cmd := parts[0] - args := parts[1:] + rt := al.buildCommandsRuntime(agent) + executor := commands.NewExecutor(al.cmdRegistry, rt) - switch cmd { - case "/show": - if len(args) < 1 { - return "Usage: /show [model|channel|agents]", true - } - switch args[0] { - case "model": - defaultAgent := al.registry.GetDefaultAgent() - if defaultAgent == nil { - return "No default agent configured", true - } - return fmt.Sprintf("Current model: %s", defaultAgent.Model), true - case "channel": - return fmt.Sprintf("Current channel: %s", msg.Channel), true - case "agents": - agentIDs := al.registry.ListAgentIDs() - return fmt.Sprintf("Registered agents: %s", strings.Join(agentIDs, ", ")), true - default: - return fmt.Sprintf("Unknown show target: %s", args[0]), true - } + var commandReply string + result := executor.Execute(ctx, commands.Request{ + Channel: msg.Channel, + ChatID: msg.ChatID, + SenderID: msg.SenderID, + Text: msg.Content, + Reply: func(text string) error { + commandReply = text + return nil + }, + }) - case "/list": - if len(args) < 1 { - return "Usage: /list [models|channels|agents]", true + switch result.Outcome { + case commands.OutcomeHandled: + if result.Err != nil { + return mapCommandError(result), true } - switch args[0] { - case "models": - return "Available models: configured in config.json per agent", true - case "channels": + if commandReply != "" { + return commandReply, true + } + return "", true + default: // OutcomePassthrough — let the message fall through to LLM + return "", false + } +} + +func (al *AgentLoop) buildCommandsRuntime(agent *AgentInstance) *commands.Runtime { + rt := &commands.Runtime{ + Config: al.cfg, + ListAgentIDs: al.registry.ListAgentIDs, + ListDefinitions: al.cmdRegistry.Definitions, + GetEnabledChannels: func() []string { if al.channelManager == nil { - return "Channel manager not initialized", true + return nil } - channels := al.channelManager.GetEnabledChannels() - if len(channels) == 0 { - return "No channels enabled", true - } - return fmt.Sprintf("Enabled channels: %s", strings.Join(channels, ", ")), true - case "agents": - agentIDs := al.registry.ListAgentIDs() - return fmt.Sprintf("Registered agents: %s", strings.Join(agentIDs, ", ")), true - default: - return fmt.Sprintf("Unknown list target: %s", args[0]), true - } - - case "/switch": - if len(args) < 3 || args[1] != "to" { - return "Usage: /switch [model|channel] to ", true - } - target := args[0] - value := args[2] - - switch target { - case "model": - defaultAgent := al.registry.GetDefaultAgent() - if defaultAgent == nil { - return "No default agent configured", true - } - oldModel := defaultAgent.Model - defaultAgent.Model = value - return fmt.Sprintf("Switched model from %s to %s", oldModel, value), true - case "channel": + return al.channelManager.GetEnabledChannels() + }, + SwitchChannel: func(value string) error { if al.channelManager == nil { - return "Channel manager not initialized", true + return fmt.Errorf("channel manager not initialized") } if _, exists := al.channelManager.GetChannel(value); !exists && value != "cli" { - return fmt.Sprintf("Channel '%s' not found or not enabled", value), true + return fmt.Errorf("channel '%s' not found or not enabled", value) } - return fmt.Sprintf("Switched target channel to %s", value), true - default: - return fmt.Sprintf("Unknown switch target: %s", target), true + return nil + }, + } + if agent != nil { + rt.GetModelInfo = func() (string, string) { + return agent.Model, al.cfg.Agents.Defaults.Provider + } + rt.SwitchModel = func(value string) (string, error) { + oldModel := agent.Model + agent.Model = value + return oldModel, nil } } + return rt +} - return "", false +func mapCommandError(result commands.ExecuteResult) string { + if result.Command == "" { + return fmt.Sprintf("Failed to execute command: %v", result.Err) + } + return fmt.Sprintf("Failed to execute /%s: %v", result.Command, result.Err) } // extractPeer extracts the routing peer from the inbound message's structured Peer field. @@ -1487,10 +1667,17 @@ func extractPeer(msg bus.InboundMessage) *routing.RoutePeer { return &routing.RoutePeer{Kind: msg.Peer.Kind, ID: peerID} } +func inboundMetadata(msg bus.InboundMessage, key string) string { + if msg.Metadata == nil { + return "" + } + return msg.Metadata[key] +} + // extractParentPeer extracts the parent peer (reply-to) from inbound message metadata. func extractParentPeer(msg bus.InboundMessage) *routing.RoutePeer { - parentKind := msg.Metadata["parent_peer_kind"] - parentID := msg.Metadata["parent_peer_id"] + parentKind := inboundMetadata(msg, metadataKeyParentPeerKind) + parentID := inboundMetadata(msg, metadataKeyParentPeerID) if parentKind == "" || parentID == "" { return nil } diff --git a/pkg/agent/loop_test.go b/pkg/agent/loop_test.go index 023286f02..2e456fa60 100644 --- a/pkg/agent/loop_test.go +++ b/pkg/agent/loop_test.go @@ -15,6 +15,7 @@ import ( "github.com/sipeed/picoclaw/pkg/config" "github.com/sipeed/picoclaw/pkg/media" "github.com/sipeed/picoclaw/pkg/providers" + "github.com/sipeed/picoclaw/pkg/routing" "github.com/sipeed/picoclaw/pkg/tools" ) @@ -164,35 +165,21 @@ func TestToolRegistry_ToolRegistration(t *testing.T) { } } -// TestToolContext_Updates verifies tool context is updated with channel/chatID +// TestToolContext_Updates verifies tool context helpers work correctly func TestToolContext_Updates(t *testing.T) { - tmpDir, err := os.MkdirTemp("", "agent-test-*") - if err != nil { - t.Fatalf("Failed to create temp dir: %v", err) - } - defer os.RemoveAll(tmpDir) + ctx := tools.WithToolContext(context.Background(), "telegram", "chat-42") - cfg := &config.Config{ - Agents: config.AgentsConfig{ - Defaults: config.AgentDefaults{ - Workspace: tmpDir, - Model: "test-model", - MaxTokens: 4096, - MaxToolIterations: 10, - }, - }, + if got := tools.ToolChannel(ctx); got != "telegram" { + t.Errorf("expected channel 'telegram', got %q", got) + } + if got := tools.ToolChatID(ctx); got != "chat-42" { + t.Errorf("expected chatID 'chat-42', got %q", got) } - msgBus := bus.NewMessageBus() - provider := &simpleMockProvider{response: "OK"} - _ = NewAgentLoop(cfg, msgBus, provider) - - // Verify that ContextualTool interface is defined and can be implemented - // This test validates the interface contract exists - ctxTool := &mockContextualTool{} - - // Verify the tool implements the interface correctly - var _ tools.ContextualTool = ctxTool + // Empty context returns empty strings + if got := tools.ToolChannel(context.Background()); got != "" { + t.Errorf("expected empty channel from bare context, got %q", got) + } } // TestToolRegistry_GetDefinitions verifies tool definitions can be retrieved @@ -241,16 +228,11 @@ func TestAgentLoop_GetStartupInfo(t *testing.T) { } defer os.RemoveAll(tmpDir) - cfg := &config.Config{ - Agents: config.AgentsConfig{ - Defaults: config.AgentDefaults{ - Workspace: tmpDir, - Model: "test-model", - MaxTokens: 4096, - MaxToolIterations: 10, - }, - }, - } + cfg := config.DefaultConfig() + cfg.Agents.Defaults.Workspace = tmpDir + cfg.Agents.Defaults.Model = "test-model" + cfg.Agents.Defaults.MaxTokens = 4096 + cfg.Agents.Defaults.MaxToolIterations = 10 msgBus := bus.NewMessageBus() provider := &mockProvider{} @@ -337,6 +319,29 @@ func (m *simpleMockProvider) GetDefaultModel() string { return "mock-model" } +type countingMockProvider struct { + response string + calls int +} + +func (m *countingMockProvider) Chat( + ctx context.Context, + messages []providers.Message, + tools []providers.ToolDefinition, + model string, + opts map[string]any, +) (*providers.LLMResponse, error) { + m.calls++ + return &providers.LLMResponse{ + Content: m.response, + ToolCalls: []providers.ToolCall{}, + }, nil +} + +func (m *countingMockProvider) GetDefaultModel() string { + return "counting-mock-model" +} + // mockCustomTool is a simple mock tool for registration testing type mockCustomTool struct{} @@ -359,36 +364,6 @@ func (m *mockCustomTool) Execute(ctx context.Context, args map[string]any) *tool return tools.SilentResult("Custom tool executed") } -// mockContextualTool tracks context updates -type mockContextualTool struct { - lastChannel string - lastChatID string -} - -func (m *mockContextualTool) Name() string { - return "mock_contextual" -} - -func (m *mockContextualTool) Description() string { - return "Mock contextual tool" -} - -func (m *mockContextualTool) Parameters() map[string]any { - return map[string]any{ - "type": "object", - "properties": map[string]any{}, - } -} - -func (m *mockContextualTool) Execute(ctx context.Context, args map[string]any) *tools.ToolResult { - return tools.SilentResult("Contextual tool executed") -} - -func (m *mockContextualTool) SetContext(channel, chatID string) { - m.lastChannel = channel - m.lastChatID = chatID -} - // testHelper executes a message and returns the response type testHelper struct { al *AgentLoop @@ -408,6 +383,198 @@ func (h testHelper) executeAndGetResponse(tb testing.TB, ctx context.Context, ms const responseTimeout = 3 * time.Second +func TestProcessMessage_UsesRouteSessionKey(t *testing.T) { + tmpDir, err := os.MkdirTemp("", "agent-test-*") + if err != nil { + t.Fatalf("Failed to create temp dir: %v", err) + } + defer os.RemoveAll(tmpDir) + + cfg := &config.Config{ + Agents: config.AgentsConfig{ + Defaults: config.AgentDefaults{ + Workspace: tmpDir, + Model: "test-model", + MaxTokens: 4096, + MaxToolIterations: 10, + }, + }, + } + + msgBus := bus.NewMessageBus() + provider := &simpleMockProvider{response: "ok"} + al := NewAgentLoop(cfg, msgBus, provider) + + msg := bus.InboundMessage{ + Channel: "telegram", + SenderID: "user1", + ChatID: "chat1", + Content: "hello", + Peer: bus.Peer{ + Kind: "direct", + ID: "user1", + }, + } + + route := al.registry.ResolveRoute(routing.RouteInput{ + Channel: msg.Channel, + Peer: extractPeer(msg), + }) + sessionKey := route.SessionKey + + defaultAgent := al.registry.GetDefaultAgent() + if defaultAgent == nil { + t.Fatal("No default agent found") + } + + helper := testHelper{al: al} + _ = helper.executeAndGetResponse(t, context.Background(), msg) + + history := defaultAgent.Sessions.GetHistory(sessionKey) + if len(history) != 2 { + t.Fatalf("expected session history len=2, got %d", len(history)) + } + if history[0].Role != "user" || history[0].Content != "hello" { + t.Fatalf("unexpected first message in session: %+v", history[0]) + } +} + +func TestProcessMessage_CommandOutcomes(t *testing.T) { + tmpDir, err := os.MkdirTemp("", "agent-test-*") + if err != nil { + t.Fatalf("Failed to create temp dir: %v", err) + } + defer os.RemoveAll(tmpDir) + + cfg := &config.Config{ + Agents: config.AgentsConfig{ + Defaults: config.AgentDefaults{ + Workspace: tmpDir, + Model: "test-model", + MaxTokens: 4096, + MaxToolIterations: 10, + }, + }, + Session: config.SessionConfig{ + DMScope: "per-channel-peer", + }, + } + + msgBus := bus.NewMessageBus() + provider := &countingMockProvider{response: "LLM reply"} + al := NewAgentLoop(cfg, msgBus, provider) + helper := testHelper{al: al} + + baseMsg := bus.InboundMessage{ + Channel: "whatsapp", + SenderID: "user1", + ChatID: "chat1", + Peer: bus.Peer{ + Kind: "direct", + ID: "user1", + }, + } + + showResp := helper.executeAndGetResponse(t, context.Background(), bus.InboundMessage{ + Channel: baseMsg.Channel, + SenderID: baseMsg.SenderID, + ChatID: baseMsg.ChatID, + Content: "/show channel", + Peer: baseMsg.Peer, + }) + if showResp != "Current Channel: whatsapp" { + t.Fatalf("unexpected /show reply: %q", showResp) + } + if provider.calls != 0 { + t.Fatalf("LLM should not be called for handled command, calls=%d", provider.calls) + } + + fooResp := helper.executeAndGetResponse(t, context.Background(), bus.InboundMessage{ + Channel: baseMsg.Channel, + SenderID: baseMsg.SenderID, + ChatID: baseMsg.ChatID, + Content: "/foo", + Peer: baseMsg.Peer, + }) + if fooResp != "LLM reply" { + t.Fatalf("unexpected /foo reply: %q", fooResp) + } + if provider.calls != 1 { + t.Fatalf("LLM should be called exactly once after /foo passthrough, calls=%d", provider.calls) + } + + newResp := helper.executeAndGetResponse(t, context.Background(), bus.InboundMessage{ + Channel: baseMsg.Channel, + SenderID: baseMsg.SenderID, + ChatID: baseMsg.ChatID, + Content: "/new", + Peer: baseMsg.Peer, + }) + if newResp != "LLM reply" { + t.Fatalf("unexpected /new reply: %q", newResp) + } + if provider.calls != 2 { + t.Fatalf("LLM should be called for passthrough /new command, calls=%d", provider.calls) + } +} + +func TestProcessMessage_SwitchModelShowModelConsistency(t *testing.T) { + tmpDir, err := os.MkdirTemp("", "agent-test-*") + if err != nil { + t.Fatalf("Failed to create temp dir: %v", err) + } + defer os.RemoveAll(tmpDir) + + cfg := &config.Config{ + Agents: config.AgentsConfig{ + Defaults: config.AgentDefaults{ + Workspace: tmpDir, + Provider: "openai", + Model: "before-switch", + MaxTokens: 4096, + MaxToolIterations: 10, + }, + }, + } + + msgBus := bus.NewMessageBus() + provider := &countingMockProvider{response: "LLM reply"} + al := NewAgentLoop(cfg, msgBus, provider) + helper := testHelper{al: al} + + switchResp := helper.executeAndGetResponse(t, context.Background(), bus.InboundMessage{ + Channel: "telegram", + SenderID: "user1", + ChatID: "chat1", + Content: "/switch model to after-switch", + Peer: bus.Peer{ + Kind: "direct", + ID: "user1", + }, + }) + if !strings.Contains(switchResp, "Switched model from before-switch to after-switch") { + t.Fatalf("unexpected /switch reply: %q", switchResp) + } + + showResp := helper.executeAndGetResponse(t, context.Background(), bus.InboundMessage{ + Channel: "telegram", + SenderID: "user1", + ChatID: "chat1", + Content: "/show model", + Peer: bus.Peer{ + Kind: "direct", + ID: "user1", + }, + }) + if !strings.Contains(showResp, "Current Model: after-switch (Provider: openai)") { + t.Fatalf("unexpected /show model reply after switch: %q", showResp) + } + + if provider.calls != 0 { + t.Fatalf("LLM should not be called for /switch and /show, calls=%d", provider.calls) + } +} + // TestToolResult_SilentToolDoesNotSendUserMessage verifies silent tools don't trigger outbound func TestToolResult_SilentToolDoesNotSendUserMessage(t *testing.T) { tmpDir, err := os.MkdirTemp("", "agent-test-*") diff --git a/pkg/agent/registry.go b/pkg/agent/registry.go index 77b846832..0e7973dc3 100644 --- a/pkg/agent/registry.go +++ b/pkg/agent/registry.go @@ -7,6 +7,7 @@ import ( "github.com/sipeed/picoclaw/pkg/logger" "github.com/sipeed/picoclaw/pkg/providers" "github.com/sipeed/picoclaw/pkg/routing" + "github.com/sipeed/picoclaw/pkg/tools" ) // AgentRegistry manages multiple agent instances and routes messages to them. @@ -100,6 +101,19 @@ func (r *AgentRegistry) CanSpawnSubagent(parentAgentID, targetAgentID string) bo return false } +// ForEachTool calls fn for every tool registered under the given name +// across all agents. This is useful for propagating dependencies (e.g. +// MediaStore) to tools after registry construction. +func (r *AgentRegistry) ForEachTool(name string, fn func(tools.Tool)) { + r.mu.RLock() + defer r.mu.RUnlock() + for _, agent := range r.agents { + if t, ok := agent.Tools.Get(name); ok { + fn(t) + } + } +} + // GetDefaultAgent returns the default agent instance. func (r *AgentRegistry) GetDefaultAgent() *AgentInstance { r.mu.RLock() diff --git a/pkg/agent/thinking.go b/pkg/agent/thinking.go new file mode 100644 index 000000000..015b69282 --- /dev/null +++ b/pkg/agent/thinking.go @@ -0,0 +1,39 @@ +package agent + +import "strings" + +// ThinkingLevel controls how the provider sends thinking parameters. +// +// - "adaptive": sends {thinking: {type: "adaptive"}} + output_config.effort (Claude 4.6+) +// - "low"/"medium"/"high"/"xhigh": sends {thinking: {type: "enabled", budget_tokens: N}} (all models) +// - "off": disables thinking +type ThinkingLevel string + +const ( + ThinkingOff ThinkingLevel = "off" + ThinkingLow ThinkingLevel = "low" + ThinkingMedium ThinkingLevel = "medium" + ThinkingHigh ThinkingLevel = "high" + ThinkingXHigh ThinkingLevel = "xhigh" + ThinkingAdaptive ThinkingLevel = "adaptive" +) + +// parseThinkingLevel normalizes a config string to a ThinkingLevel. +// Case-insensitive and whitespace-tolerant for user-facing config values. +// Returns ThinkingOff for unknown or empty values. +func parseThinkingLevel(level string) ThinkingLevel { + switch strings.ToLower(strings.TrimSpace(level)) { + case "adaptive": + return ThinkingAdaptive + case "low": + return ThinkingLow + case "medium": + return ThinkingMedium + case "high": + return ThinkingHigh + case "xhigh": + return ThinkingXHigh + default: + return ThinkingOff + } +} diff --git a/pkg/agent/thinking_test.go b/pkg/agent/thinking_test.go new file mode 100644 index 000000000..be3a68c33 --- /dev/null +++ b/pkg/agent/thinking_test.go @@ -0,0 +1,35 @@ +package agent + +import "testing" + +func TestParseThinkingLevel(t *testing.T) { + tests := []struct { + name string + input string + want ThinkingLevel + }{ + {"off", "off", ThinkingOff}, + {"empty", "", ThinkingOff}, + {"low", "low", ThinkingLow}, + {"medium", "medium", ThinkingMedium}, + {"high", "high", ThinkingHigh}, + {"xhigh", "xhigh", ThinkingXHigh}, + {"adaptive", "adaptive", ThinkingAdaptive}, + {"unknown", "unknown", ThinkingOff}, + // Case-insensitive and whitespace-tolerant + {"upper_Medium", "Medium", ThinkingMedium}, + {"upper_HIGH", "HIGH", ThinkingHigh}, + {"mixed_Adaptive", "Adaptive", ThinkingAdaptive}, + {"leading_space", " high", ThinkingHigh}, + {"trailing_space", "low ", ThinkingLow}, + {"both_spaces", " medium ", ThinkingMedium}, + } + + for _, tt := range tests { + t.Run(tt.name, func(t *testing.T) { + if got := parseThinkingLevel(tt.input); got != tt.want { + t.Errorf("parseThinkingLevel(%q) = %q, want %q", tt.input, got, tt.want) + } + }) + } +} diff --git a/pkg/auth/anthropic_usage.go b/pkg/auth/anthropic_usage.go new file mode 100644 index 000000000..716b2908e --- /dev/null +++ b/pkg/auth/anthropic_usage.go @@ -0,0 +1,71 @@ +package auth + +import ( + "encoding/json" + "fmt" + "io" + "net/http" + "time" +) + +const ( + anthropicBetaHeader = "oauth-2025-04-20" + anthropicAPIVersion = "2023-06-01" +) + +// anthropicUsageURL is the endpoint for fetching OAuth usage stats. +// It is a var (not const) to allow overriding in tests. +var anthropicUsageURL = "https://api.anthropic.com/api/oauth/usage" + +func setAnthropicUsageURL(url string) { anthropicUsageURL = url } + +type AnthropicUsage struct { + FiveHourUtilization float64 + SevenDayUtilization float64 +} + +func FetchAnthropicUsage(token string) (*AnthropicUsage, error) { + req, err := http.NewRequest("GET", anthropicUsageURL, nil) + if err != nil { + return nil, err + } + req.Header.Set("Authorization", "Bearer "+token) + req.Header.Set("Anthropic-Version", anthropicAPIVersion) + req.Header.Set("Anthropic-Beta", anthropicBetaHeader) + + client := &http.Client{Timeout: 10 * time.Second} + resp, err := client.Do(req) + if err != nil { + return nil, err + } + defer resp.Body.Close() + + body, err := io.ReadAll(resp.Body) + if err != nil { + return nil, fmt.Errorf("reading usage response: %w", err) + } + + if resp.StatusCode != http.StatusOK { + if resp.StatusCode == http.StatusForbidden { + return nil, fmt.Errorf("insufficient scope: usage endpoint requires oauth scope") + } + return nil, fmt.Errorf("usage request failed (%d): %s", resp.StatusCode, string(body)) + } + + var result struct { + FiveHour struct { + Utilization float64 `json:"utilization"` + } `json:"five_hour"` + SevenDay struct { + Utilization float64 `json:"utilization"` + } `json:"seven_day"` + } + if err := json.Unmarshal(body, &result); err != nil { + return nil, fmt.Errorf("parsing usage response: %w", err) + } + + return &AnthropicUsage{ + FiveHourUtilization: result.FiveHour.Utilization, + SevenDayUtilization: result.SevenDay.Utilization, + }, nil +} diff --git a/pkg/auth/anthropic_usage_test.go b/pkg/auth/anthropic_usage_test.go new file mode 100644 index 000000000..ef4a35364 --- /dev/null +++ b/pkg/auth/anthropic_usage_test.go @@ -0,0 +1,98 @@ +package auth + +import ( + "net/http" + "net/http/httptest" + "strings" + "testing" +) + +func TestFetchAnthropicUsage_Success(t *testing.T) { + srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + if got := r.Header.Get("Authorization"); got != "Bearer test-token" { + t.Errorf("Authorization = %q, want %q", got, "Bearer test-token") + } + if got := r.Header.Get("Anthropic-Beta"); got != anthropicBetaHeader { + t.Errorf("Anthropic-Beta = %q, want %q", got, anthropicBetaHeader) + } + w.WriteHeader(http.StatusOK) + w.Write([]byte(`{"five_hour":{"utilization":0.42},"seven_day":{"utilization":0.85}}`)) + })) + defer srv.Close() + + // Temporarily override the URL by using the test server + origURL := anthropicUsageURL + defer func() { setAnthropicUsageURL(origURL) }() + setAnthropicUsageURL(srv.URL) + + usage, err := FetchAnthropicUsage("test-token") + if err != nil { + t.Fatalf("unexpected error: %v", err) + } + if usage.FiveHourUtilization != 0.42 { + t.Errorf("FiveHourUtilization = %v, want 0.42", usage.FiveHourUtilization) + } + if usage.SevenDayUtilization != 0.85 { + t.Errorf("SevenDayUtilization = %v, want 0.85", usage.SevenDayUtilization) + } +} + +func TestFetchAnthropicUsage_Forbidden(t *testing.T) { + srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + w.WriteHeader(http.StatusForbidden) + w.Write([]byte(`{"error":"forbidden"}`)) + })) + defer srv.Close() + + origURL := anthropicUsageURL + defer func() { setAnthropicUsageURL(origURL) }() + setAnthropicUsageURL(srv.URL) + + _, err := FetchAnthropicUsage("test-token") + if err == nil { + t.Fatal("expected error for 403, got nil") + } + if !strings.Contains(err.Error(), "insufficient scope") { + t.Errorf("expected 'insufficient scope' error, got %q", err.Error()) + } +} + +func TestFetchAnthropicUsage_ServerError(t *testing.T) { + srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + w.WriteHeader(http.StatusInternalServerError) + w.Write([]byte(`internal error`)) + })) + defer srv.Close() + + origURL := anthropicUsageURL + defer func() { setAnthropicUsageURL(origURL) }() + setAnthropicUsageURL(srv.URL) + + _, err := FetchAnthropicUsage("test-token") + if err == nil { + t.Fatal("expected error for 500, got nil") + } + if !strings.Contains(err.Error(), "500") { + t.Errorf("expected error containing '500', got %q", err.Error()) + } +} + +func TestFetchAnthropicUsage_MalformedJSON(t *testing.T) { + srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + w.WriteHeader(http.StatusOK) + w.Write([]byte(`not json`)) + })) + defer srv.Close() + + origURL := anthropicUsageURL + defer func() { setAnthropicUsageURL(origURL) }() + setAnthropicUsageURL(srv.URL) + + _, err := FetchAnthropicUsage("test-token") + if err == nil { + t.Fatal("expected error for malformed JSON, got nil") + } + if !strings.Contains(err.Error(), "parsing usage response") { + t.Errorf("expected 'parsing usage response' error, got %q", err.Error()) + } +} diff --git a/pkg/auth/oauth.go b/pkg/auth/oauth.go index 91c9e25c5..4667e3d81 100644 --- a/pkg/auth/oauth.go +++ b/pkg/auth/oauth.go @@ -212,7 +212,10 @@ func RequestDeviceCode(cfg OAuthProviderConfig) (*DeviceCodeInfo, error) { } defer resp.Body.Close() - body, _ := io.ReadAll(resp.Body) + body, err := io.ReadAll(resp.Body) + if err != nil { + return nil, fmt.Errorf("reading device code response: %w", err) + } if resp.StatusCode != http.StatusOK { return nil, fmt.Errorf("device code request failed: %s", string(body)) } @@ -300,7 +303,10 @@ func LoginDeviceCode(cfg OAuthProviderConfig) (*AuthCredential, error) { } defer resp.Body.Close() - body, _ := io.ReadAll(resp.Body) + body, err := io.ReadAll(resp.Body) + if err != nil { + return nil, fmt.Errorf("reading device code response: %w", err) + } if resp.StatusCode != http.StatusOK { return nil, fmt.Errorf("device code request failed: %s", string(body)) } @@ -360,7 +366,10 @@ func pollDeviceCode(cfg OAuthProviderConfig, deviceAuthID, userCode string) (*Au return nil, fmt.Errorf("pending") } - body, _ := io.ReadAll(resp.Body) + body, err := io.ReadAll(resp.Body) + if err != nil { + return nil, fmt.Errorf("reading device token response: %w", err) + } var tokenResp struct { AuthorizationCode string `json:"authorization_code"` @@ -401,7 +410,10 @@ func RefreshAccessToken(cred *AuthCredential, cfg OAuthProviderConfig) (*AuthCre } defer resp.Body.Close() - body, _ := io.ReadAll(resp.Body) + body, err := io.ReadAll(resp.Body) + if err != nil { + return nil, fmt.Errorf("reading token refresh response: %w", err) + } if resp.StatusCode != http.StatusOK { return nil, fmt.Errorf("token refresh failed: %s", string(body)) } @@ -494,7 +506,10 @@ func ExchangeCodeForTokens(cfg OAuthProviderConfig, code, codeVerifier, redirect } defer resp.Body.Close() - body, _ := io.ReadAll(resp.Body) + body, err := io.ReadAll(resp.Body) + if err != nil { + return nil, fmt.Errorf("reading token exchange response: %w", err) + } if resp.StatusCode != http.StatusOK { return nil, fmt.Errorf("token exchange failed: %s", string(body)) } diff --git a/pkg/auth/store.go b/pkg/auth/store.go index 283dc6977..2e55d4877 100644 --- a/pkg/auth/store.go +++ b/pkg/auth/store.go @@ -39,6 +39,9 @@ func (c *AuthCredential) NeedsRefresh() bool { } func authFilePath() string { + if home := os.Getenv("PICOCLAW_HOME"); home != "" { + return filepath.Join(home, "auth.json") + } home, _ := os.UserHomeDir() return filepath.Join(home, ".picoclaw", "auth.json") } diff --git a/pkg/auth/token.go b/pkg/auth/token.go index a5a13ff03..0e69e60ac 100644 --- a/pkg/auth/token.go +++ b/pkg/auth/token.go @@ -31,6 +31,35 @@ func LoginPasteToken(provider string, r io.Reader) (*AuthCredential, error) { }, nil } +func LoginSetupToken(r io.Reader) (*AuthCredential, error) { + fmt.Println("Paste your setup token from `claude setup-token`:") + fmt.Print("> ") + + scanner := bufio.NewScanner(r) + if !scanner.Scan() { + if err := scanner.Err(); err != nil { + return nil, fmt.Errorf("reading token: %w", err) + } + return nil, fmt.Errorf("no input received") + } + + token := strings.TrimSpace(scanner.Text()) + + if !strings.HasPrefix(token, "sk-ant-oat01-") { + return nil, fmt.Errorf("invalid setup token: expected prefix sk-ant-oat01-") + } + + if len(token) < 80 { + return nil, fmt.Errorf("invalid setup token: too short (expected at least 80 characters)") + } + + return &AuthCredential{ + AccessToken: token, + Provider: "anthropic", + AuthMethod: "oauth", + }, nil +} + func providerDisplayName(provider string) string { switch provider { case "anthropic": diff --git a/pkg/auth/token_test.go b/pkg/auth/token_test.go new file mode 100644 index 000000000..673cd9d5d --- /dev/null +++ b/pkg/auth/token_test.go @@ -0,0 +1,61 @@ +package auth + +import ( + "strings" + "testing" +) + +func TestLoginSetupToken(t *testing.T) { + // A valid token: correct prefix + at least 80 chars + validToken := "sk-ant-oat01-" + strings.Repeat("a", 80) + + tests := []struct { + name string + input string + wantErr string + }{ + {"valid token", validToken, ""}, + {"empty input", "", "expected prefix sk-ant-oat01-"}, + {"wrong prefix", "sk-ant-api-" + strings.Repeat("a", 80), "expected prefix sk-ant-oat01-"}, + {"too short", "sk-ant-oat01-short", "too short"}, + {"whitespace only", " ", "expected prefix sk-ant-oat01-"}, + } + + for _, tt := range tests { + t.Run(tt.name, func(t *testing.T) { + r := strings.NewReader(tt.input + "\n") + cred, err := LoginSetupToken(r) + + if tt.wantErr != "" { + if err == nil { + t.Fatalf("expected error containing %q, got nil", tt.wantErr) + } + if !strings.Contains(err.Error(), tt.wantErr) { + t.Fatalf("expected error containing %q, got %q", tt.wantErr, err.Error()) + } + return + } + + if err != nil { + t.Fatalf("unexpected error: %v", err) + } + if cred.AccessToken != validToken { + t.Errorf("AccessToken = %q, want %q", cred.AccessToken, validToken) + } + if cred.Provider != "anthropic" { + t.Errorf("Provider = %q, want %q", cred.Provider, "anthropic") + } + if cred.AuthMethod != "oauth" { + t.Errorf("AuthMethod = %q, want %q", cred.AuthMethod, "oauth") + } + }) + } +} + +func TestLoginSetupToken_EmptyReader(t *testing.T) { + r := strings.NewReader("") + _, err := LoginSetupToken(r) + if err == nil { + t.Fatal("expected error for empty reader, got nil") + } +} diff --git a/pkg/channels/discord/discord.go b/pkg/channels/discord/discord.go index cd6a2560f..c3bcbff8d 100644 --- a/pkg/channels/discord/discord.go +++ b/pkg/channels/discord/discord.go @@ -3,12 +3,16 @@ package discord import ( "context" "fmt" + "net/http" + "net/url" "os" + "regexp" "strings" "sync" "time" "github.com/bwmarrin/discordgo" + "github.com/gorilla/websocket" "github.com/sipeed/picoclaw/pkg/bus" "github.com/sipeed/picoclaw/pkg/channels" @@ -23,6 +27,12 @@ const ( sendTimeout = 10 * time.Second ) +var ( + // Pre-compiled regexes for resolveDiscordRefs (avoid re-compiling per call) + channelRefRe = regexp.MustCompile(`<#(\d+)>`) + msgLinkRe = regexp.MustCompile(`https://(?:discord\.com|discordapp\.com)/channels/(\d+)/(\d+)/(\d+)`) +) + type DiscordChannel struct { *channels.BaseChannel session *discordgo.Session @@ -40,6 +50,9 @@ func NewDiscordChannel(cfg config.DiscordConfig, bus *bus.MessageBus) (*DiscordC return nil, fmt.Errorf("failed to create discord session: %w", err) } + if err := applyDiscordProxy(session, cfg.Proxy); err != nil { + return nil, err + } base := channels.NewBaseChannel("discord", cfg, bus, cfg.AllowFrom, channels.WithMaxMessageLength(2000), channels.WithGroupTrigger(cfg.GroupTrigger), @@ -332,6 +345,24 @@ func (c *DiscordChannel) handleMessage(s *discordgo.Session, m *discordgo.Messag content = c.stripBotMention(content) } + // Resolve Discord refs in main content before concatenation to avoid + // double-expanding links that appear in the referenced message. + content = c.resolveDiscordRefs(s, content, m.GuildID) + + // Prepend referenced (quoted) message content if this is a reply + if m.MessageReference != nil && m.ReferencedMessage != nil { + refContent := m.ReferencedMessage.Content + if refContent != "" { + refAuthor := "unknown" + if m.ReferencedMessage.Author != nil { + refAuthor = m.ReferencedMessage.Author.Username + } + refContent = c.resolveDiscordRefs(s, refContent, m.GuildID) + content = fmt.Sprintf("[quoted message from %s]: %s\n\n%s", + refAuthor, refContent, content) + } + } + senderID := m.Author.ID mediaPaths := make([]string, 0, len(m.Attachments)) @@ -465,9 +496,88 @@ func (c *DiscordChannel) StartTyping(ctx context.Context, chatID string) (func() func (c *DiscordChannel) downloadAttachment(url, filename string) string { return utils.DownloadFile(url, filename, utils.DownloadOptions{ LoggerPrefix: "discord", + ProxyURL: c.config.Proxy, }) } +func applyDiscordProxy(session *discordgo.Session, proxyAddr string) error { + var proxyFunc func(*http.Request) (*url.URL, error) + if proxyAddr != "" { + proxyURL, err := url.Parse(proxyAddr) + if err != nil { + return fmt.Errorf("invalid discord proxy URL %q: %w", proxyAddr, err) + } + proxyFunc = http.ProxyURL(proxyURL) + } else if os.Getenv("HTTP_PROXY") != "" || os.Getenv("HTTPS_PROXY") != "" { + proxyFunc = http.ProxyFromEnvironment + } + + if proxyFunc == nil { + return nil + } + + transport := &http.Transport{Proxy: proxyFunc} + session.Client = &http.Client{ + Timeout: sendTimeout, + Transport: transport, + } + + if session.Dialer != nil { + dialerCopy := *session.Dialer + dialerCopy.Proxy = proxyFunc + session.Dialer = &dialerCopy + } else { + session.Dialer = &websocket.Dialer{Proxy: proxyFunc} + } + + return nil +} + +// resolveDiscordRefs resolves channel references (<#id> → #channel-name) and +// expands Discord message links to show the linked message content. +// Only links pointing to the same guild are expanded to prevent cross-guild leakage. +func (c *DiscordChannel) resolveDiscordRefs(s *discordgo.Session, text string, guildID string) string { + // 1. Resolve channel references: <#id> → #channel-name + text = channelRefRe.ReplaceAllStringFunc(text, func(match string) string { + parts := channelRefRe.FindStringSubmatch(match) + if len(parts) < 2 { + return match + } + // Prefer session state cache to avoid API calls + if ch, err := s.State.Channel(parts[1]); err == nil { + return "#" + ch.Name + } + if ch, err := s.Channel(parts[1]); err == nil { + return "#" + ch.Name + } + return match + }) + + // 2. Expand Discord message links (max 3, same guild only) + matches := msgLinkRe.FindAllStringSubmatch(text, 3) + for _, m := range matches { + if len(m) < 4 { + continue + } + linkGuildID, channelID, messageID := m[1], m[2], m[3] + // Security: only expand links from the same guild + if linkGuildID != guildID { + continue + } + msg, err := s.ChannelMessage(channelID, messageID) + if err != nil || msg == nil || msg.Content == "" { + continue + } + author := "unknown" + if msg.Author != nil { + author = msg.Author.Username + } + text += fmt.Sprintf("\n[linked message from %s]: %s", author, msg.Content) + } + + return text +} + // stripBotMention removes the bot mention from the message content. // Discord mentions have the format <@USER_ID> or <@!USER_ID> (with nickname). func (c *DiscordChannel) stripBotMention(text string) string { diff --git a/pkg/channels/discord/discord_resolve_test.go b/pkg/channels/discord/discord_resolve_test.go new file mode 100644 index 000000000..4bc65cc18 --- /dev/null +++ b/pkg/channels/discord/discord_resolve_test.go @@ -0,0 +1,98 @@ +package discord + +import ( + "testing" +) + +func TestChannelRefRegex(t *testing.T) { + tests := []struct { + name string + input string + wantID string + wantOK bool + }{ + {"basic channel ref", "<#123456789>", "123456789", true}, + {"long id", "<#9876543210123456>", "9876543210123456", true}, + {"no match plain text", "hello world", "", false}, + {"no match partial", "<#>", "", false}, + {"no match letters", "<#abc>", "", false}, + } + + for _, tt := range tests { + t.Run(tt.name, func(t *testing.T) { + matches := channelRefRe.FindStringSubmatch(tt.input) + if tt.wantOK { + if len(matches) < 2 || matches[1] != tt.wantID { + t.Errorf("channelRefRe(%q) = %v, want ID %q", tt.input, matches, tt.wantID) + } + } else { + if len(matches) >= 2 { + t.Errorf("channelRefRe(%q) should not match, got %v", tt.input, matches) + } + } + }) + } +} + +func TestMsgLinkRegex(t *testing.T) { + tests := []struct { + name string + input string + wantGuild string + wantChan string + wantMsg string + wantOK bool + }{ + { + "discord.com link", + "https://discord.com/channels/111/222/333", + "111", "222", "333", true, + }, + { + "discordapp.com link", + "https://discordapp.com/channels/111/222/333", + "111", "222", "333", true, + }, + { + "real world ids", + "check this https://discord.com/channels/9000000000000001/9000000000000002/9000000000000003 please", + "9000000000000001", "9000000000000002", "9000000000000003", true, + }, + {"no match http", "http://discord.com/channels/1/2/3", "", "", "", false}, + {"no match missing segment", "https://discord.com/channels/1/2", "", "", "", false}, + {"no match plain text", "hello world", "", "", "", false}, + } + + for _, tt := range tests { + t.Run(tt.name, func(t *testing.T) { + matches := msgLinkRe.FindStringSubmatch(tt.input) + if tt.wantOK { + if len(matches) < 4 { + t.Fatalf("msgLinkRe(%q) didn't match, want guild=%s chan=%s msg=%s", + tt.input, tt.wantGuild, tt.wantChan, tt.wantMsg) + } + if matches[1] != tt.wantGuild || matches[2] != tt.wantChan || matches[3] != tt.wantMsg { + t.Errorf("msgLinkRe(%q) = guild=%s chan=%s msg=%s, want %s/%s/%s", + tt.input, matches[1], matches[2], matches[3], + tt.wantGuild, tt.wantChan, tt.wantMsg) + } + } else { + if len(matches) >= 4 { + t.Errorf("msgLinkRe(%q) should not match, got %v", tt.input, matches) + } + } + }) + } +} + +func TestMsgLinkRegex_MultipleMatches(t *testing.T) { + input := "see https://discord.com/channels/1/2/3 and https://discord.com/channels/4/5/6 and https://discord.com/channels/7/8/9 and https://discord.com/channels/10/11/12" + matches := msgLinkRe.FindAllStringSubmatch(input, 3) + if len(matches) != 3 { + t.Fatalf("expected 3 matches (capped), got %d", len(matches)) + } + // Verify the 3rd match is 7/8/9 (not 10/11/12) + if matches[2][1] != "7" || matches[2][2] != "8" || matches[2][3] != "9" { + t.Errorf("3rd match = %v, want guild=7 chan=8 msg=9", matches[2]) + } +} diff --git a/pkg/channels/discord/discord_test.go b/pkg/channels/discord/discord_test.go new file mode 100644 index 000000000..0cd5328f4 --- /dev/null +++ b/pkg/channels/discord/discord_test.go @@ -0,0 +1,91 @@ +package discord + +import ( + "net/http" + "net/url" + "testing" + + "github.com/bwmarrin/discordgo" +) + +func TestApplyDiscordProxy_CustomProxy(t *testing.T) { + session, err := discordgo.New("Bot test-token") + if err != nil { + t.Fatalf("discordgo.New() error: %v", err) + } + + if err = applyDiscordProxy(session, "http://127.0.0.1:7890"); err != nil { + t.Fatalf("applyDiscordProxy() error: %v", err) + } + + req, err := http.NewRequest("GET", "https://discord.com/api/v10/gateway", nil) + if err != nil { + t.Fatalf("http.NewRequest() error: %v", err) + } + + restProxy := session.Client.Transport.(*http.Transport).Proxy + restProxyURL, err := restProxy(req) + if err != nil { + t.Fatalf("rest proxy func error: %v", err) + } + if got, want := restProxyURL.String(), "http://127.0.0.1:7890"; got != want { + t.Fatalf("REST proxy = %q, want %q", got, want) + } + + wsProxyURL, err := session.Dialer.Proxy(req) + if err != nil { + t.Fatalf("ws proxy func error: %v", err) + } + if got, want := wsProxyURL.String(), "http://127.0.0.1:7890"; got != want { + t.Fatalf("WS proxy = %q, want %q", got, want) + } +} + +func TestApplyDiscordProxy_FromEnvironment(t *testing.T) { + t.Setenv("HTTP_PROXY", "http://127.0.0.1:8888") + t.Setenv("http_proxy", "http://127.0.0.1:8888") + t.Setenv("HTTPS_PROXY", "http://127.0.0.1:8888") + t.Setenv("https_proxy", "http://127.0.0.1:8888") + t.Setenv("ALL_PROXY", "") + t.Setenv("all_proxy", "") + t.Setenv("NO_PROXY", "") + t.Setenv("no_proxy", "") + + session, err := discordgo.New("Bot test-token") + if err != nil { + t.Fatalf("discordgo.New() error: %v", err) + } + + if err = applyDiscordProxy(session, ""); err != nil { + t.Fatalf("applyDiscordProxy() error: %v", err) + } + + req, err := http.NewRequest("GET", "https://discord.com/api/v10/gateway", nil) + if err != nil { + t.Fatalf("http.NewRequest() error: %v", err) + } + + gotURL, err := session.Dialer.Proxy(req) + if err != nil { + t.Fatalf("ws proxy func error: %v", err) + } + + wantURL, err := url.Parse("http://127.0.0.1:8888") + if err != nil { + t.Fatalf("url.Parse() error: %v", err) + } + if gotURL.String() != wantURL.String() { + t.Fatalf("WS proxy = %q, want %q", gotURL.String(), wantURL.String()) + } +} + +func TestApplyDiscordProxy_InvalidProxyURL(t *testing.T) { + session, err := discordgo.New("Bot test-token") + if err != nil { + t.Fatalf("discordgo.New() error: %v", err) + } + + if err = applyDiscordProxy(session, "://bad-proxy"); err == nil { + t.Fatal("applyDiscordProxy() expected error for invalid proxy URL, got nil") + } +} diff --git a/pkg/channels/interfaces.go b/pkg/channels/interfaces.go index 74caeeac5..b3a493761 100644 --- a/pkg/channels/interfaces.go +++ b/pkg/channels/interfaces.go @@ -1,6 +1,10 @@ package channels -import "context" +import ( + "context" + + "github.com/sipeed/picoclaw/pkg/commands" +) // TypingCapable — channels that can show a typing/thinking indicator. // StartTyping begins the indicator and returns a stop function. @@ -39,3 +43,10 @@ type PlaceholderRecorder interface { RecordTypingStop(channel, chatID string, stop func()) RecordReactionUndo(channel, chatID string, undo func()) } + +// CommandRegistrarCapable is implemented by channels that can register +// command menus with their upstream platform (e.g. Telegram BotCommand). +// Channels that do not support platform-level command menus can ignore it. +type CommandRegistrarCapable interface { + RegisterCommands(ctx context.Context, defs []commands.Definition) error +} diff --git a/pkg/channels/interfaces_command_test.go b/pkg/channels/interfaces_command_test.go new file mode 100644 index 000000000..de5502644 --- /dev/null +++ b/pkg/channels/interfaces_command_test.go @@ -0,0 +1,16 @@ +package channels + +import ( + "context" + "testing" + + "github.com/sipeed/picoclaw/pkg/commands" +) + +type mockRegistrar struct{} + +func (mockRegistrar) RegisterCommands(context.Context, []commands.Definition) error { return nil } + +func TestCommandRegistrarCapable_Compiles(t *testing.T) { + var _ CommandRegistrarCapable = mockRegistrar{} +} diff --git a/pkg/channels/line/line.go b/pkg/channels/line/line.go index 398f12e6b..b36350a06 100644 --- a/pkg/channels/line/line.go +++ b/pkg/channels/line/line.go @@ -654,7 +654,10 @@ func (c *LINEChannel) callAPI(ctx context.Context, endpoint string, payload any) defer resp.Body.Close() if resp.StatusCode != http.StatusOK { - respBody, _ := io.ReadAll(resp.Body) + respBody, err := io.ReadAll(resp.Body) + if err != nil { + return channels.ClassifySendError(resp.StatusCode, fmt.Errorf("reading LINE API error response: %w", err)) + } return channels.ClassifySendError(resp.StatusCode, fmt.Errorf("LINE API error: %s", string(respBody))) } diff --git a/pkg/channels/telegram/command_registration.go b/pkg/channels/telegram/command_registration.go new file mode 100644 index 000000000..d3152ec3d --- /dev/null +++ b/pkg/channels/telegram/command_registration.go @@ -0,0 +1,116 @@ +package telegram + +import ( + "context" + "math/rand" + "slices" + "time" + + "github.com/mymmrac/telego" + + "github.com/sipeed/picoclaw/pkg/commands" + "github.com/sipeed/picoclaw/pkg/logger" +) + +var commandRegistrationBackoff = []time.Duration{ + 5 * time.Second, + 15 * time.Second, + 60 * time.Second, + 5 * time.Minute, + 10 * time.Minute, +} + +func commandRegistrationDelay(attempt int) time.Duration { + if len(commandRegistrationBackoff) == 0 { + return 0 + } + base := commandRegistrationBackoff[min(attempt, len(commandRegistrationBackoff)-1)] + // Full jitter in [0.5, 1.0) to avoid synchronized retries across instances. + return time.Duration(float64(base) * (0.5 + rand.Float64()*0.5)) +} + +// RegisterCommands registers bot commands on Telegram platform. +func (c *TelegramChannel) RegisterCommands(ctx context.Context, defs []commands.Definition) error { + botCommands := make([]telego.BotCommand, 0, len(defs)) + for _, def := range defs { + if def.Name == "" || def.Description == "" { + continue + } + botCommands = append(botCommands, telego.BotCommand{ + Command: def.Name, + Description: def.Description, + }) + } + + current, err := c.bot.GetMyCommands(ctx, &telego.GetMyCommandsParams{}) + if err != nil { + // If we can't read current commands, fall through to set them. + logger.WarnCF("telegram", "Failed to get current commands, will set unconditionally", + map[string]any{"error": err.Error()}) + } else if slices.Equal(current, botCommands) { + logger.DebugCF("telegram", "Bot commands are up to date", nil) + return nil + } + + return c.bot.SetMyCommands(ctx, &telego.SetMyCommandsParams{ + Commands: botCommands, + }) +} + +func (c *TelegramChannel) startCommandRegistration(ctx context.Context, defs []commands.Definition) { + if len(defs) == 0 { + return + } + + register := c.registerFunc + if register == nil { + register = c.RegisterCommands + } + + regCtx, cancel := context.WithCancel(ctx) + c.commandRegCancel = cancel + + // Registration runs asynchronously so Telegram message intake is never blocked + // by temporary upstream API failures. Retry stops on success or channel shutdown. + go func() { + attempt := 0 + timer := time.NewTimer(0) + if !timer.Stop() { + select { + case <-timer.C: + default: + } + } + defer timer.Stop() + for { + err := register(regCtx, defs) + if err == nil { + logger.InfoCF("telegram", "Telegram commands registered", map[string]any{ + "count": len(defs), + }) + return + } + + delay := commandRegistrationDelay(attempt) + logger.WarnCF("telegram", "Telegram command registration failed; will retry", map[string]any{ + "error": err.Error(), + "retry_after": delay.String(), + }) + attempt++ + + if !timer.Stop() { + select { + case <-timer.C: + default: + } + } + timer.Reset(delay) + + select { + case <-regCtx.Done(): + return + case <-timer.C: + } + } + }() +} diff --git a/pkg/channels/telegram/command_registration_test.go b/pkg/channels/telegram/command_registration_test.go new file mode 100644 index 000000000..26f891b2e --- /dev/null +++ b/pkg/channels/telegram/command_registration_test.go @@ -0,0 +1,96 @@ +package telegram + +import ( + "context" + "errors" + "sync/atomic" + "testing" + "time" + + "github.com/sipeed/picoclaw/pkg/commands" +) + +func TestStartCommandRegistration_DoesNotBlock(t *testing.T) { + ch := &TelegramChannel{} + started := make(chan struct{}, 1) + ctx, cancel := context.WithCancel(context.Background()) + defer cancel() + + ch.registerFunc = func(context.Context, []commands.Definition) error { + started <- struct{}{} + return errors.New("temporary failure") + } + + ch.startCommandRegistration(ctx, []commands.Definition{{Name: "help"}}) + + select { + case <-started: + case <-time.After(time.Second): + t.Fatal("registration did not start asynchronously") + } +} + +func TestStartCommandRegistration_RetriesUntilSuccessThenStops(t *testing.T) { + ch := &TelegramChannel{} + ctx, cancel := context.WithCancel(context.Background()) + defer cancel() + + origBackoff := commandRegistrationBackoff + commandRegistrationBackoff = []time.Duration{5 * time.Millisecond} + defer func() { commandRegistrationBackoff = origBackoff }() + + var attempts atomic.Int32 + ch.registerFunc = func(context.Context, []commands.Definition) error { + n := attempts.Add(1) + if n < 3 { + return errors.New("temporary failure") + } + return nil + } + + ch.startCommandRegistration(ctx, []commands.Definition{{Name: "help", Description: "Help"}}) + + deadline := time.Now().Add(250 * time.Millisecond) + for time.Now().Before(deadline) { + if attempts.Load() >= 3 { + break + } + time.Sleep(5 * time.Millisecond) + } + if attempts.Load() < 3 { + t.Fatalf("expected at least 3 attempts, got %d", attempts.Load()) + } + + stable := attempts.Load() + time.Sleep(30 * time.Millisecond) + if attempts.Load() != stable { + t.Fatalf("expected retries to stop after success, got %d -> %d", stable, attempts.Load()) + } +} + +func TestStartCommandRegistration_StopsAfterCancel(t *testing.T) { + ch := &TelegramChannel{} + ctx, cancel := context.WithCancel(context.Background()) + + origBackoff := commandRegistrationBackoff + commandRegistrationBackoff = []time.Duration{5 * time.Millisecond} + defer func() { commandRegistrationBackoff = origBackoff }() + defer cancel() + + var attempts atomic.Int32 + ch.registerFunc = func(context.Context, []commands.Definition) error { + attempts.Add(1) + return errors.New("always fail") + } + + ch.startCommandRegistration(ctx, []commands.Definition{{Name: "help", Description: "Help"}}) + + time.Sleep(20 * time.Millisecond) + cancel() + time.Sleep(20 * time.Millisecond) // allow in-flight attempt to settle + stable := attempts.Load() + time.Sleep(30 * time.Millisecond) + if attempts.Load() != stable { + t.Fatalf("expected retries to quiesce after cancel, got %d -> %d", stable, attempts.Load()) + } +} diff --git a/pkg/channels/telegram/telegram.go b/pkg/channels/telegram/telegram.go index f328f32b8..a2035853c 100644 --- a/pkg/channels/telegram/telegram.go +++ b/pkg/channels/telegram/telegram.go @@ -7,7 +7,6 @@ import ( "net/url" "os" "regexp" - "slices" "strconv" "strings" "time" @@ -18,6 +17,7 @@ import ( "github.com/sipeed/picoclaw/pkg/bus" "github.com/sipeed/picoclaw/pkg/channels" + "github.com/sipeed/picoclaw/pkg/commands" "github.com/sipeed/picoclaw/pkg/config" "github.com/sipeed/picoclaw/pkg/identity" "github.com/sipeed/picoclaw/pkg/logger" @@ -40,13 +40,15 @@ var ( type TelegramChannel struct { *channels.BaseChannel - bot *telego.Bot - bh *th.BotHandler - commands TelegramCommander - config *config.Config - chatIDs map[string]int64 - ctx context.Context - cancel context.CancelFunc + bot *telego.Bot + bh *th.BotHandler + config *config.Config + chatIDs map[string]int64 + ctx context.Context + cancel context.CancelFunc + + registerFunc func(context.Context, []commands.Definition) error + commandRegCancel context.CancelFunc } func NewTelegramChannel(cfg *config.Config, bus *bus.MessageBus) (*TelegramChannel, error) { @@ -93,7 +95,6 @@ func NewTelegramChannel(cfg *config.Config, bus *bus.MessageBus) (*TelegramChann return &TelegramChannel{ BaseChannel: base, - commands: NewTelegramCommands(bot, cfg), bot: bot, config: cfg, chatIDs: make(map[string]int64), @@ -105,12 +106,6 @@ func (c *TelegramChannel) Start(ctx context.Context) error { c.ctx, c.cancel = context.WithCancel(ctx) - if err := c.initBotCommands(c.ctx); err != nil { - logger.WarnCF("telegram", "Failed to initialize bot commands", map[string]any{ - "error": err.Error(), - }) - } - updates, err := c.bot.UpdatesViaLongPolling(c.ctx, &telego.GetUpdatesParams{ Timeout: 30, }) @@ -126,21 +121,6 @@ func (c *TelegramChannel) Start(ctx context.Context) error { } c.bh = bh - bh.HandleMessage(func(ctx *th.Context, message telego.Message) error { - return c.commands.Start(ctx, message) - }, th.CommandEqual("start")) - bh.HandleMessage(func(ctx *th.Context, message telego.Message) error { - return c.commands.Help(ctx, message) - }, th.CommandEqual("help")) - - bh.HandleMessage(func(ctx *th.Context, message telego.Message) error { - return c.commands.Show(ctx, message) - }, th.CommandEqual("show")) - - bh.HandleMessage(func(ctx *th.Context, message telego.Message) error { - return c.commands.List(ctx, message) - }, th.CommandEqual("list")) - bh.HandleMessage(func(ctx *th.Context, message telego.Message) error { return c.handleMessage(ctx, &message) }, th.AnyMessage()) @@ -150,6 +130,8 @@ func (c *TelegramChannel) Start(ctx context.Context) error { "username": c.bot.Username(), }) + c.startCommandRegistration(c.ctx, commands.BuiltinDefinitions()) + go func() { if err = bh.Start(); err != nil { logger.ErrorCF("telegram", "Bot handler failed", map[string]any{ @@ -174,50 +156,8 @@ func (c *TelegramChannel) Stop(ctx context.Context) error { if c.cancel != nil { c.cancel() } - - return nil -} - -func (c *TelegramChannel) initBotCommands(ctx context.Context) error { - currentCommands, err := c.bot.GetMyCommands(ctx, &telego.GetMyCommandsParams{ - Scope: tu.ScopeDefault(), - }) - if err != nil { - return fmt.Errorf("get commands: %w", err) - } - - commands := []telego.BotCommand{ - { - Command: "start", - Description: "Start the bot", - }, - { - Command: "help", - Description: "Show a help message", - }, - { - Command: "show", - Description: "Show current configuration", - }, - { - Command: "list", - Description: "List available options", - }, - } - - // Setting commands on each start will hit the rate limit very quickly, that's why we check if an update is needed - if !slices.Equal(currentCommands, commands) { - logger.InfoC("telegram", "Updating bot commands") - - err = c.bot.SetMyCommands(ctx, &telego.SetMyCommandsParams{ - Commands: commands, - Scope: tu.ScopeDefault(), - }) - if err != nil { - return fmt.Errorf("set commands: %w", err) - } - } else { - logger.DebugC("telegram", "Bot commands are up to date") + if c.commandRegCancel != nil { + c.commandRegCancel() } return nil @@ -721,34 +661,34 @@ func escapeHTML(text string) string { // isBotMentioned checks if the bot is mentioned in the message via entities. func (c *TelegramChannel) isBotMentioned(message *telego.Message) bool { - botUsername := c.bot.Username() - if botUsername == "" { + text, entities := telegramEntityTextAndList(message) + if text == "" || len(entities) == 0 { return false } - entities := message.Entities - if entities == nil { - entities = message.CaptionEntities + botUsername := "" + if c.bot != nil { + botUsername = c.bot.Username() } + runes := []rune(text) for _, entity := range entities { - if entity.Type == "mention" { - // Extract the mention text from the message - text := message.Text - if text == "" { - text = message.Caption - } - runes := []rune(text) - end := entity.Offset + entity.Length - if end <= len(runes) { - mention := string(runes[entity.Offset:end]) - if strings.EqualFold(mention, "@"+botUsername) { - return true - } - } + entityText, ok := telegramEntityText(runes, entity) + if !ok { + continue } - if entity.Type == "text_mention" && entity.User != nil { - if entity.User.Username == botUsername { + + switch entity.Type { + case telego.EntityTypeMention: + if botUsername != "" && strings.EqualFold(entityText, "@"+botUsername) { + return true + } + case telego.EntityTypeTextMention: + if botUsername != "" && entity.User != nil && strings.EqualFold(entity.User.Username, botUsername) { + return true + } + case telego.EntityTypeBotCommand: + if isBotCommandEntityForThisBot(entityText, botUsername) { return true } } @@ -756,6 +696,46 @@ func (c *TelegramChannel) isBotMentioned(message *telego.Message) bool { return false } +func telegramEntityTextAndList(message *telego.Message) (string, []telego.MessageEntity) { + if message.Text != "" { + return message.Text, message.Entities + } + return message.Caption, message.CaptionEntities +} + +func telegramEntityText(runes []rune, entity telego.MessageEntity) (string, bool) { + if entity.Offset < 0 || entity.Length <= 0 { + return "", false + } + end := entity.Offset + entity.Length + if entity.Offset >= len(runes) || end > len(runes) { + return "", false + } + return string(runes[entity.Offset:end]), true +} + +func isBotCommandEntityForThisBot(entityText, botUsername string) bool { + if !strings.HasPrefix(entityText, "/") { + return false + } + command := strings.TrimPrefix(entityText, "/") + if command == "" { + return false + } + + at := strings.IndexRune(command, '@') + if at == -1 { + // A bare /command delivered to this bot is intended for this bot. + return true + } + + mentionUsername := command[at+1:] + if mentionUsername == "" || botUsername == "" { + return false + } + return strings.EqualFold(mentionUsername, botUsername) +} + // stripBotMention removes the @bot mention from the content. func (c *TelegramChannel) stripBotMention(content string) string { botUsername := c.bot.Username() diff --git a/pkg/channels/telegram/telegram_commands.go b/pkg/channels/telegram/telegram_commands.go deleted file mode 100644 index 496fc5e4f..000000000 --- a/pkg/channels/telegram/telegram_commands.go +++ /dev/null @@ -1,156 +0,0 @@ -package telegram - -import ( - "context" - "fmt" - "strings" - - "github.com/mymmrac/telego" - - "github.com/sipeed/picoclaw/pkg/config" -) - -type TelegramCommander interface { - Help(ctx context.Context, message telego.Message) error - Start(ctx context.Context, message telego.Message) error - Show(ctx context.Context, message telego.Message) error - List(ctx context.Context, message telego.Message) error -} - -type cmd struct { - bot *telego.Bot - config *config.Config -} - -func NewTelegramCommands(bot *telego.Bot, cfg *config.Config) TelegramCommander { - return &cmd{ - bot: bot, - config: cfg, - } -} - -func commandArgs(text string) string { - parts := strings.SplitN(text, " ", 2) - if len(parts) < 2 { - return "" - } - return strings.TrimSpace(parts[1]) -} - -func (c *cmd) Help(ctx context.Context, message telego.Message) error { - msg := `/start - Start the bot -/help - Show this help message -/show [model|channel] - Show current configuration -/list [models|channels] - List available options - ` - _, err := c.bot.SendMessage(ctx, &telego.SendMessageParams{ - ChatID: telego.ChatID{ID: message.Chat.ID}, - Text: msg, - ReplyParameters: &telego.ReplyParameters{ - MessageID: message.MessageID, - }, - }) - return err -} - -func (c *cmd) Start(ctx context.Context, message telego.Message) error { - _, err := c.bot.SendMessage(ctx, &telego.SendMessageParams{ - ChatID: telego.ChatID{ID: message.Chat.ID}, - Text: "Hello! I am PicoClaw 🦞", - ReplyParameters: &telego.ReplyParameters{ - MessageID: message.MessageID, - }, - }) - return err -} - -func (c *cmd) Show(ctx context.Context, message telego.Message) error { - args := commandArgs(message.Text) - if args == "" { - _, err := c.bot.SendMessage(ctx, &telego.SendMessageParams{ - ChatID: telego.ChatID{ID: message.Chat.ID}, - Text: "Usage: /show [model|channel]", - ReplyParameters: &telego.ReplyParameters{ - MessageID: message.MessageID, - }, - }) - return err - } - - var response string - switch args { - case "model": - response = fmt.Sprintf("Current Model: %s (Provider: %s)", - c.config.Agents.Defaults.GetModelName(), - c.config.Agents.Defaults.Provider) - case "channel": - response = "Current Channel: telegram" - default: - response = fmt.Sprintf("Unknown parameter: %s. Try 'model' or 'channel'.", args) - } - - _, err := c.bot.SendMessage(ctx, &telego.SendMessageParams{ - ChatID: telego.ChatID{ID: message.Chat.ID}, - Text: response, - ReplyParameters: &telego.ReplyParameters{ - MessageID: message.MessageID, - }, - }) - return err -} - -func (c *cmd) List(ctx context.Context, message telego.Message) error { - args := commandArgs(message.Text) - if args == "" { - _, err := c.bot.SendMessage(ctx, &telego.SendMessageParams{ - ChatID: telego.ChatID{ID: message.Chat.ID}, - Text: "Usage: /list [models|channels]", - ReplyParameters: &telego.ReplyParameters{ - MessageID: message.MessageID, - }, - }) - return err - } - - var response string - switch args { - case "models": - provider := c.config.Agents.Defaults.Provider - if provider == "" { - provider = "configured default" - } - response = fmt.Sprintf("Configured Model: %s\nProvider: %s\n\nTo change models, update config.json", - c.config.Agents.Defaults.GetModelName(), provider) - - case "channels": - var enabled []string - if c.config.Channels.Telegram.Enabled { - enabled = append(enabled, "telegram") - } - if c.config.Channels.WhatsApp.Enabled { - enabled = append(enabled, "whatsapp") - } - if c.config.Channels.Feishu.Enabled { - enabled = append(enabled, "feishu") - } - if c.config.Channels.Discord.Enabled { - enabled = append(enabled, "discord") - } - if c.config.Channels.Slack.Enabled { - enabled = append(enabled, "slack") - } - response = fmt.Sprintf("Enabled Channels:\n- %s", strings.Join(enabled, "\n- ")) - - default: - response = fmt.Sprintf("Unknown parameter: %s. Try 'models' or 'channels'.", args) - } - - _, err := c.bot.SendMessage(ctx, &telego.SendMessageParams{ - ChatID: telego.ChatID{ID: message.Chat.ID}, - Text: response, - ReplyParameters: &telego.ReplyParameters{ - MessageID: message.MessageID, - }, - }) - return err -} diff --git a/pkg/channels/telegram/telegram_dispatch_test.go b/pkg/channels/telegram/telegram_dispatch_test.go new file mode 100644 index 000000000..1ea4a4824 --- /dev/null +++ b/pkg/channels/telegram/telegram_dispatch_test.go @@ -0,0 +1,52 @@ +package telegram + +import ( + "context" + "testing" + "time" + + "github.com/mymmrac/telego" + + "github.com/sipeed/picoclaw/pkg/bus" + "github.com/sipeed/picoclaw/pkg/channels" +) + +func TestHandleMessage_DoesNotConsumeGenericCommandsLocally(t *testing.T) { + messageBus := bus.NewMessageBus() + ch := &TelegramChannel{ + BaseChannel: channels.NewBaseChannel("telegram", nil, messageBus, nil), + chatIDs: make(map[string]int64), + ctx: context.Background(), + } + + msg := &telego.Message{ + Text: "/new", + MessageID: 9, + Chat: telego.Chat{ + ID: 123, + Type: "private", + }, + From: &telego.User{ + ID: 42, + FirstName: "Alice", + }, + } + + if err := ch.handleMessage(context.Background(), msg); err != nil { + t.Fatalf("handleMessage error: %v", err) + } + + ctx, cancel := context.WithTimeout(context.Background(), time.Second) + defer cancel() + + inbound, ok := messageBus.ConsumeInbound(ctx) + if !ok { + t.Fatal("expected inbound message to be forwarded") + } + if inbound.Channel != "telegram" { + t.Fatalf("channel=%q", inbound.Channel) + } + if inbound.Content != "/new" { + t.Fatalf("content=%q", inbound.Content) + } +} diff --git a/pkg/channels/telegram/telegram_group_command_filter_test.go b/pkg/channels/telegram/telegram_group_command_filter_test.go new file mode 100644 index 000000000..0d5b985fe --- /dev/null +++ b/pkg/channels/telegram/telegram_group_command_filter_test.go @@ -0,0 +1,147 @@ +package telegram + +import ( + "context" + "fmt" + "strings" + "testing" + "time" + + "github.com/mymmrac/telego" + ta "github.com/mymmrac/telego/telegoapi" + + "github.com/sipeed/picoclaw/pkg/bus" + "github.com/sipeed/picoclaw/pkg/channels" + "github.com/sipeed/picoclaw/pkg/config" +) + +type getMeCaller struct { + username string +} + +func (c getMeCaller) Call(_ context.Context, url string, _ *ta.RequestData) (*ta.Response, error) { + if strings.HasSuffix(url, "/getMe") { + result := fmt.Sprintf(`{"id":1,"is_bot":true,"first_name":"bot","username":%q}`, c.username) + return &ta.Response{Ok: true, Result: []byte(result)}, nil + } + return &ta.Response{Ok: true, Result: []byte("true")}, nil +} + +func newTestTelegramBot(t *testing.T, username string) *telego.Bot { + t.Helper() + + token := "123456:" + strings.Repeat("a", 35) + bot, err := telego.NewBot(token, + telego.WithAPICaller(getMeCaller{username: username}), + telego.WithDiscardLogger(), + ) + if err != nil { + t.Fatalf("NewBot error: %v", err) + } + return bot +} + +func newGroupMentionOnlyChannel(t *testing.T, botUsername string) (*TelegramChannel, *bus.MessageBus) { + t.Helper() + + messageBus := bus.NewMessageBus() + ch := &TelegramChannel{ + BaseChannel: channels.NewBaseChannel("telegram", nil, messageBus, nil, + channels.WithGroupTrigger(config.GroupTriggerConfig{MentionOnly: true}), + ), + bot: newTestTelegramBot(t, botUsername), + chatIDs: make(map[string]int64), + ctx: context.Background(), + } + return ch, messageBus +} + +func TestHandleMessage_GroupMentionOnly_BotCommandEntity(t *testing.T) { + tests := []struct { + name string + text string + wantForwarded bool + wantContent string + }{ + { + name: "command with bot username", + text: "/new@testbot", + wantForwarded: true, + wantContent: "/new", + }, + { + name: "bare command", + text: "/new", + wantForwarded: true, + wantContent: "/new", + }, + { + name: "command for another bot", + text: "/new@otherbot", + wantForwarded: false, + }, + } + + for _, tc := range tests { + t.Run(tc.name, func(t *testing.T) { + ch, messageBus := newGroupMentionOnlyChannel(t, "testbot") + + msg := &telego.Message{ + Text: tc.text, + Entities: []telego.MessageEntity{{ + Type: telego.EntityTypeBotCommand, + Offset: 0, + Length: len([]rune(tc.text)), + }}, + MessageID: 42, + Chat: telego.Chat{ + ID: 123, + Type: "group", + }, + From: &telego.User{ + ID: 7, + FirstName: "Alice", + }, + } + + if err := ch.handleMessage(context.Background(), msg); err != nil { + t.Fatalf("handleMessage error: %v", err) + } + + ctx, cancel := context.WithTimeout(context.Background(), 150*time.Millisecond) + defer cancel() + + inbound, ok := messageBus.ConsumeInbound(ctx) + if tc.wantForwarded { + if !ok { + t.Fatal("expected inbound message to be forwarded") + } + if inbound.Content != tc.wantContent { + t.Fatalf("content=%q want=%q", inbound.Content, tc.wantContent) + } + return + } + + if ok { + t.Fatalf("expected message to be filtered, got content=%q", inbound.Content) + } + }) + } +} + +func TestIsBotMentioned_MentionEntityUnaffected(t *testing.T) { + ch, _ := newGroupMentionOnlyChannel(t, "testbot") + + msg := &telego.Message{ + Text: "@testbot hello", + Entities: []telego.MessageEntity{{ + Type: telego.EntityTypeMention, + Offset: 0, + Length: len("@testbot"), + }}, + } + + if !ch.isBotMentioned(msg) { + t.Fatal("expected mention entity to be treated as bot mention") + } +} diff --git a/pkg/channels/wecom/aibot.go b/pkg/channels/wecom/aibot.go index 6c5aca40b..93fe8c36d 100644 --- a/pkg/channels/wecom/aibot.go +++ b/pkg/channels/wecom/aibot.go @@ -793,7 +793,10 @@ func (c *WeComAIBotChannel) sendViaResponseURL(responseURL, content string) erro return nil } - respBody, _ := io.ReadAll(resp.Body) + respBody, err := io.ReadAll(resp.Body) + if err != nil { + return fmt.Errorf("reading response_url body: %w: %w", channels.ErrTemporary, err) + } switch { case resp.StatusCode == http.StatusTooManyRequests: return fmt.Errorf("response_url rate limited (%d): %s: %w", diff --git a/pkg/channels/wecom/app.go b/pkg/channels/wecom/app.go index 717815b9f..2098fcd4e 100644 --- a/pkg/channels/wecom/app.go +++ b/pkg/channels/wecom/app.go @@ -321,8 +321,17 @@ func (c *WeComAppChannel) uploadMedia(ctx context.Context, accessToken, mediaTyp defer resp.Body.Close() if resp.StatusCode != http.StatusOK { - respBody, _ := io.ReadAll(resp.Body) - return "", channels.ClassifySendError(resp.StatusCode, fmt.Errorf("wecom upload error: %s", string(respBody))) + respBody, readErr := io.ReadAll(resp.Body) + if readErr != nil { + return "", channels.ClassifySendError( + resp.StatusCode, + fmt.Errorf("reading wecom upload error response: %w", readErr), + ) + } + return "", channels.ClassifySendError( + resp.StatusCode, + fmt.Errorf("wecom upload error: %s", string(respBody)), + ) } var result struct { @@ -371,8 +380,17 @@ func (c *WeComAppChannel) sendWeComMessage(ctx context.Context, accessToken stri defer resp.Body.Close() if resp.StatusCode != http.StatusOK { - respBody, _ := io.ReadAll(resp.Body) - return channels.ClassifySendError(resp.StatusCode, fmt.Errorf("wecom_app API error: %s", string(respBody))) + respBody, readErr := io.ReadAll(resp.Body) + if readErr != nil { + return channels.ClassifySendError( + resp.StatusCode, + fmt.Errorf("reading wecom_app error response: %w", readErr), + ) + } + return channels.ClassifySendError( + resp.StatusCode, + fmt.Errorf("wecom_app API error: %s", string(respBody)), + ) } respBody, err := io.ReadAll(resp.Body) diff --git a/pkg/channels/wecom/bot.go b/pkg/channels/wecom/bot.go index 9126a847d..96d5a961f 100644 --- a/pkg/channels/wecom/bot.go +++ b/pkg/channels/wecom/bot.go @@ -453,8 +453,17 @@ func (c *WeComBotChannel) sendWebhookReply(ctx context.Context, userID, content defer resp.Body.Close() if resp.StatusCode != http.StatusOK { - body, _ := io.ReadAll(resp.Body) - return channels.ClassifySendError(resp.StatusCode, fmt.Errorf("webhook API error: %s", string(body))) + body, readErr := io.ReadAll(resp.Body) + if readErr != nil { + return channels.ClassifySendError( + resp.StatusCode, + fmt.Errorf("reading webhook error response: %w", readErr), + ) + } + return channels.ClassifySendError( + resp.StatusCode, + fmt.Errorf("webhook API error: %s", string(body)), + ) } body, err := io.ReadAll(resp.Body) diff --git a/pkg/channels/whatsapp/whatsapp_command_test.go b/pkg/channels/whatsapp/whatsapp_command_test.go new file mode 100644 index 000000000..ee8aa4a52 --- /dev/null +++ b/pkg/channels/whatsapp/whatsapp_command_test.go @@ -0,0 +1,41 @@ +package whatsapp + +import ( + "context" + "testing" + "time" + + "github.com/sipeed/picoclaw/pkg/bus" + "github.com/sipeed/picoclaw/pkg/channels" + "github.com/sipeed/picoclaw/pkg/config" +) + +func TestHandleIncomingMessage_DoesNotConsumeGenericCommandsLocally(t *testing.T) { + messageBus := bus.NewMessageBus() + ch := &WhatsAppChannel{ + BaseChannel: channels.NewBaseChannel("whatsapp", config.WhatsAppConfig{}, messageBus, nil), + ctx: context.Background(), + } + + ch.handleIncomingMessage(map[string]any{ + "type": "message", + "id": "mid1", + "from": "user1", + "chat": "chat1", + "content": "/help", + }) + + ctx, cancel := context.WithTimeout(context.Background(), time.Second) + defer cancel() + + inbound, ok := messageBus.ConsumeInbound(ctx) + if !ok { + t.Fatal("expected inbound message to be forwarded") + } + if inbound.Channel != "whatsapp" { + t.Fatalf("channel=%q", inbound.Channel) + } + if inbound.Content != "/help" { + t.Fatalf("content=%q", inbound.Content) + } +} diff --git a/pkg/channels/whatsapp_native/whatsapp_command_test.go b/pkg/channels/whatsapp_native/whatsapp_command_test.go new file mode 100644 index 000000000..cc2dcb619 --- /dev/null +++ b/pkg/channels/whatsapp_native/whatsapp_command_test.go @@ -0,0 +1,56 @@ +//go:build whatsapp_native + +package whatsapp + +import ( + "context" + "testing" + "time" + + "go.mau.fi/whatsmeow/proto/waE2E" + "go.mau.fi/whatsmeow/types" + "go.mau.fi/whatsmeow/types/events" + "google.golang.org/protobuf/proto" + + "github.com/sipeed/picoclaw/pkg/bus" + "github.com/sipeed/picoclaw/pkg/channels" + "github.com/sipeed/picoclaw/pkg/config" +) + +func TestHandleIncoming_DoesNotConsumeGenericCommandsLocally(t *testing.T) { + messageBus := bus.NewMessageBus() + ch := &WhatsAppNativeChannel{ + BaseChannel: channels.NewBaseChannel("whatsapp_native", config.WhatsAppConfig{}, messageBus, nil), + runCtx: context.Background(), + } + + evt := &events.Message{ + Info: types.MessageInfo{ + MessageSource: types.MessageSource{ + Sender: types.NewJID("1001", types.DefaultUserServer), + Chat: types.NewJID("1001", types.DefaultUserServer), + }, + ID: "mid1", + PushName: "Alice", + }, + Message: &waE2E.Message{ + Conversation: proto.String("/new"), + }, + } + + ch.handleIncoming(evt) + + ctx, cancel := context.WithTimeout(context.Background(), time.Second) + defer cancel() + + inbound, ok := messageBus.ConsumeInbound(ctx) + if !ok { + t.Fatal("expected inbound message to be forwarded") + } + if inbound.Channel != "whatsapp_native" { + t.Fatalf("channel=%q", inbound.Channel) + } + if inbound.Content != "/new" { + t.Fatalf("content=%q", inbound.Content) + } +} diff --git a/pkg/commands/builtin.go b/pkg/commands/builtin.go new file mode 100644 index 000000000..a36dd3eba --- /dev/null +++ b/pkg/commands/builtin.go @@ -0,0 +1,16 @@ +package commands + +// BuiltinDefinitions returns all built-in command definitions. +// Each command group is defined in its own cmd_*.go file. +// Definitions are stateless — runtime dependencies are provided +// via the Runtime parameter passed to handlers at execution time. +func BuiltinDefinitions() []Definition { + return []Definition{ + startCommand(), + helpCommand(), + showCommand(), + listCommand(), + switchCommand(), + checkCommand(), + } +} diff --git a/pkg/commands/builtin_test.go b/pkg/commands/builtin_test.go new file mode 100644 index 000000000..66a84825e --- /dev/null +++ b/pkg/commands/builtin_test.go @@ -0,0 +1,145 @@ +package commands + +import ( + "context" + "strings" + "testing" +) + +func findDefinitionByName(t *testing.T, defs []Definition, name string) Definition { + t.Helper() + for _, def := range defs { + if def.Name == name { + return def + } + } + t.Fatalf("missing /%s definition", name) + return Definition{} +} + +func TestBuiltinHelpHandler_ReturnsFormattedMessage(t *testing.T) { + defs := BuiltinDefinitions() + helpDef := findDefinitionByName(t, defs, "help") + if helpDef.Handler == nil { + t.Fatalf("/help handler should not be nil") + } + + var reply string + err := helpDef.Handler(context.Background(), Request{ + Text: "/help", + Reply: func(text string) error { + reply = text + return nil + }, + }, nil) + if err != nil { + t.Fatalf("/help handler error: %v", err) + } + // Now uses auto-generated EffectiveUsage which includes agents + if !strings.Contains(reply, "/show [model|channel|agents]") { + t.Fatalf("/help reply missing /show usage, got %q", reply) + } + if !strings.Contains(reply, "/list [models|channels|agents]") { + t.Fatalf("/help reply missing /list usage, got %q", reply) + } +} + +func TestBuiltinShowChannel_PreservesUserVisibleBehavior(t *testing.T) { + defs := BuiltinDefinitions() + ex := NewExecutor(NewRegistry(defs), nil) + + cases := []string{"telegram", "whatsapp"} + for _, channel := range cases { + var reply string + res := ex.Execute(context.Background(), Request{ + Channel: channel, + Text: "/show channel", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("/show channel on %s: outcome=%v, want=%v", channel, res.Outcome, OutcomeHandled) + } + want := "Current Channel: " + channel + if reply != want { + t.Fatalf("/show channel reply=%q, want=%q", reply, want) + } + } +} + +func TestBuiltinListChannels_UsesGetEnabledChannels(t *testing.T) { + rt := &Runtime{ + GetEnabledChannels: func() []string { + return []string{"telegram", "slack"} + }, + } + defs := BuiltinDefinitions() + ex := NewExecutor(NewRegistry(defs), rt) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/list channels", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("/list channels: outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if !strings.Contains(reply, "telegram") || !strings.Contains(reply, "slack") { + t.Fatalf("/list channels reply=%q, want telegram and slack", reply) + } +} + +func TestBuiltinShowAgents_RestoresOldBehavior(t *testing.T) { + rt := &Runtime{ + ListAgentIDs: func() []string { + return []string{"default", "coder"} + }, + } + defs := BuiltinDefinitions() + ex := NewExecutor(NewRegistry(defs), rt) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/show agents", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("/show agents: outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if !strings.Contains(reply, "default") || !strings.Contains(reply, "coder") { + t.Fatalf("/show agents reply=%q, want agent IDs", reply) + } +} + +func TestBuiltinListAgents_RestoresOldBehavior(t *testing.T) { + rt := &Runtime{ + ListAgentIDs: func() []string { + return []string{"default", "coder"} + }, + } + defs := BuiltinDefinitions() + ex := NewExecutor(NewRegistry(defs), rt) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/list agents", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("/list agents: outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if !strings.Contains(reply, "default") || !strings.Contains(reply, "coder") { + t.Fatalf("/list agents reply=%q, want agent IDs", reply) + } +} diff --git a/pkg/commands/cmd_check.go b/pkg/commands/cmd_check.go new file mode 100644 index 000000000..f0193dc4f --- /dev/null +++ b/pkg/commands/cmd_check.go @@ -0,0 +1,33 @@ +package commands + +import ( + "context" + "fmt" +) + +func checkCommand() Definition { + return Definition{ + Name: "check", + Description: "Check channel availability", + SubCommands: []SubCommand{ + { + Name: "channel", + Description: "Check if a channel is available", + ArgsUsage: "", + Handler: func(_ context.Context, req Request, rt *Runtime) error { + if rt == nil || rt.SwitchChannel == nil { + return req.Reply(unavailableMsg) + } + value := nthToken(req.Text, 2) + if value == "" { + return req.Reply("Usage: /check channel ") + } + if err := rt.SwitchChannel(value); err != nil { + return req.Reply(err.Error()) + } + return req.Reply(fmt.Sprintf("Channel '%s' is available and enabled", value)) + }, + }, + }, + } +} diff --git a/pkg/commands/cmd_help.go b/pkg/commands/cmd_help.go new file mode 100644 index 000000000..94f7f0101 --- /dev/null +++ b/pkg/commands/cmd_help.go @@ -0,0 +1,44 @@ +package commands + +import ( + "context" + "fmt" + "strings" +) + +func helpCommand() Definition { + return Definition{ + Name: "help", + Description: "Show this help message", + Usage: "/help", + Handler: func(_ context.Context, req Request, rt *Runtime) error { + var defs []Definition + if rt != nil && rt.ListDefinitions != nil { + defs = rt.ListDefinitions() + } else { + defs = BuiltinDefinitions() + } + return req.Reply(formatHelpMessage(defs)) + }, + } +} + +func formatHelpMessage(defs []Definition) string { + if len(defs) == 0 { + return "No commands available." + } + + lines := make([]string, 0, len(defs)) + for _, def := range defs { + usage := def.EffectiveUsage() + if usage == "" { + usage = "/" + def.Name + } + desc := def.Description + if desc == "" { + desc = "No description" + } + lines = append(lines, fmt.Sprintf("%s - %s", usage, desc)) + } + return strings.Join(lines, "\n") +} diff --git a/pkg/commands/cmd_list.go b/pkg/commands/cmd_list.go new file mode 100644 index 000000000..bf47b6e9c --- /dev/null +++ b/pkg/commands/cmd_list.go @@ -0,0 +1,52 @@ +package commands + +import ( + "context" + "fmt" + "strings" +) + +func listCommand() Definition { + return Definition{ + Name: "list", + Description: "List available options", + SubCommands: []SubCommand{ + { + Name: "models", + Description: "Configured models", + Handler: func(_ context.Context, req Request, rt *Runtime) error { + if rt == nil || rt.GetModelInfo == nil { + return req.Reply(unavailableMsg) + } + name, provider := rt.GetModelInfo() + if provider == "" { + provider = "configured default" + } + return req.Reply(fmt.Sprintf( + "Configured Model: %s\nProvider: %s\n\nTo change models, update config.json", + name, provider, + )) + }, + }, + { + Name: "channels", + Description: "Enabled channels", + Handler: func(_ context.Context, req Request, rt *Runtime) error { + if rt == nil || rt.GetEnabledChannels == nil { + return req.Reply(unavailableMsg) + } + enabled := rt.GetEnabledChannels() + if len(enabled) == 0 { + return req.Reply("No channels enabled") + } + return req.Reply(fmt.Sprintf("Enabled Channels:\n- %s", strings.Join(enabled, "\n- "))) + }, + }, + { + Name: "agents", + Description: "Registered agents", + Handler: agentsHandler(), + }, + }, + } +} diff --git a/pkg/commands/cmd_show.go b/pkg/commands/cmd_show.go new file mode 100644 index 000000000..c655e6880 --- /dev/null +++ b/pkg/commands/cmd_show.go @@ -0,0 +1,38 @@ +package commands + +import ( + "context" + "fmt" +) + +func showCommand() Definition { + return Definition{ + Name: "show", + Description: "Show current configuration", + SubCommands: []SubCommand{ + { + Name: "model", + Description: "Current model and provider", + Handler: func(_ context.Context, req Request, rt *Runtime) error { + if rt == nil || rt.GetModelInfo == nil { + return req.Reply(unavailableMsg) + } + name, provider := rt.GetModelInfo() + return req.Reply(fmt.Sprintf("Current Model: %s (Provider: %s)", name, provider)) + }, + }, + { + Name: "channel", + Description: "Current channel", + Handler: func(_ context.Context, req Request, _ *Runtime) error { + return req.Reply(fmt.Sprintf("Current Channel: %s", req.Channel)) + }, + }, + { + Name: "agents", + Description: "Registered agents", + Handler: agentsHandler(), + }, + }, + } +} diff --git a/pkg/commands/cmd_start.go b/pkg/commands/cmd_start.go new file mode 100644 index 000000000..8b500aa10 --- /dev/null +++ b/pkg/commands/cmd_start.go @@ -0,0 +1,14 @@ +package commands + +import "context" + +func startCommand() Definition { + return Definition{ + Name: "start", + Description: "Start the bot", + Usage: "/start", + Handler: func(_ context.Context, req Request, _ *Runtime) error { + return req.Reply("Hello! I am PicoClaw 🦞") + }, + } +} diff --git a/pkg/commands/cmd_switch.go b/pkg/commands/cmd_switch.go new file mode 100644 index 000000000..fb8fc109e --- /dev/null +++ b/pkg/commands/cmd_switch.go @@ -0,0 +1,42 @@ +package commands + +import ( + "context" + "fmt" +) + +func switchCommand() Definition { + return Definition{ + Name: "switch", + Description: "Switch model", + SubCommands: []SubCommand{ + { + Name: "model", + Description: "Switch to a different model", + ArgsUsage: "to ", + Handler: func(_ context.Context, req Request, rt *Runtime) error { + if rt == nil || rt.SwitchModel == nil { + return req.Reply(unavailableMsg) + } + // Parse: /switch model to + value := nthToken(req.Text, 3) // tokens: [/switch, model, to, ] + if nthToken(req.Text, 2) != "to" || value == "" { + return req.Reply("Usage: /switch model to ") + } + oldModel, err := rt.SwitchModel(value) + if err != nil { + return req.Reply(err.Error()) + } + return req.Reply(fmt.Sprintf("Switched model from %s to %s", oldModel, value)) + }, + }, + { + Name: "channel", + Description: "Moved to /check channel", + Handler: func(_ context.Context, req Request, _ *Runtime) error { + return req.Reply("This command has moved. Please use: /check channel ") + }, + }, + }, + } +} diff --git a/pkg/commands/cmd_switch_test.go b/pkg/commands/cmd_switch_test.go new file mode 100644 index 000000000..59ed305bb --- /dev/null +++ b/pkg/commands/cmd_switch_test.go @@ -0,0 +1,279 @@ +package commands + +import ( + "context" + "fmt" + "testing" +) + +func TestSwitchModel_Success(t *testing.T) { + rt := &Runtime{ + SwitchModel: func(value string) (string, error) { + return "old-model", nil + }, + } + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), rt) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/switch model to gpt-4", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + want := "Switched model from old-model to gpt-4" + if reply != want { + t.Fatalf("reply=%q, want=%q", reply, want) + } +} + +func TestSwitchModel_MissingToKeyword(t *testing.T) { + rt := &Runtime{ + SwitchModel: func(value string) (string, error) { + return "old", nil + }, + } + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), rt) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/switch model gpt-4", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if reply != "Usage: /switch model to " { + t.Fatalf("reply=%q, want usage message", reply) + } +} + +func TestSwitchModel_MissingValue(t *testing.T) { + rt := &Runtime{ + SwitchModel: func(value string) (string, error) { + return "old", nil + }, + } + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), rt) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/switch model to", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if reply != "Usage: /switch model to " { + t.Fatalf("reply=%q, want usage message", reply) + } +} + +func TestSwitchModel_Error(t *testing.T) { + rt := &Runtime{ + SwitchModel: func(value string) (string, error) { + return "", fmt.Errorf("model not found") + }, + } + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), rt) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/switch model to bad-model", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if reply != "model not found" { + t.Fatalf("reply=%q, want error message", reply) + } +} + +func TestSwitchModel_NilDep(t *testing.T) { + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), &Runtime{}) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/switch model to gpt-4", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if reply != "Command unavailable in current context." { + t.Fatalf("reply=%q, want unavailable message", reply) + } +} + +func TestSwitchChannel_Redirect(t *testing.T) { + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), &Runtime{}) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/switch channel to telegram", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + want := "This command has moved. Please use: /check channel " + if reply != want { + t.Fatalf("reply=%q, want=%q", reply, want) + } +} + +func TestCheckChannel_Success(t *testing.T) { + rt := &Runtime{ + SwitchChannel: func(value string) error { + return nil + }, + } + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), rt) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/check channel telegram", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + want := "Channel 'telegram' is available and enabled" + if reply != want { + t.Fatalf("reply=%q, want=%q", reply, want) + } +} + +func TestCheckChannel_Error(t *testing.T) { + rt := &Runtime{ + SwitchChannel: func(value string) error { + return fmt.Errorf("channel '%s' not found", value) + }, + } + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), rt) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/check channel unknown", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if reply != "channel 'unknown' not found" { + t.Fatalf("reply=%q, want error message", reply) + } +} + +func TestCheckChannel_NilDep(t *testing.T) { + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), &Runtime{}) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/check channel telegram", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if reply != "Command unavailable in current context." { + t.Fatalf("reply=%q, want unavailable message", reply) + } +} + +func TestCheckChannel_MissingValue(t *testing.T) { + rt := &Runtime{ + SwitchChannel: func(value string) error { + return nil + }, + } + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), rt) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/check channel", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if reply != "Usage: /check channel " { + t.Fatalf("reply=%q, want usage message", reply) + } +} + +func TestSwitch_BangPrefix(t *testing.T) { + rt := &Runtime{ + SwitchModel: func(value string) (string, error) { + return "old", nil + }, + } + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), rt) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "!switch model to gpt-4", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("! prefix: outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if reply != "Switched model from old to gpt-4" { + t.Fatalf("! prefix: reply=%q, want success message", reply) + } +} + +func TestSwitch_NoSubCommand(t *testing.T) { + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), &Runtime{}) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/switch", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + // Should get usage message from executor's sub-command routing + if reply == "" { + t.Fatal("expected usage reply for bare /switch") + } +} diff --git a/pkg/commands/definition.go b/pkg/commands/definition.go new file mode 100644 index 000000000..7309df317 --- /dev/null +++ b/pkg/commands/definition.go @@ -0,0 +1,48 @@ +package commands + +import ( + "fmt" + "strings" +) + +// SubCommand defines a single sub-command within a parent command. +type SubCommand struct { + Name string + Description string + ArgsUsage string // optional, e.g. "" + Handler Handler +} + +// Definition is the single-source metadata and behavior contract for a slash command. +// +// Design notes (phase 1): +// - Every channel reads command shape from this type instead of keeping local copies. +// - Visibility is global: all definitions are considered available to all channels. +// - Platform menu registration (for example Telegram BotCommand) also derives from this +// same definition so UI labels and runtime behavior stay aligned. +type Definition struct { + Name string + Description string + Usage string // for simple commands; ignored when SubCommands is set + Aliases []string + SubCommands []SubCommand // optional; when set, Executor routes to sub-command handlers + Handler Handler // for simple commands without sub-commands +} + +// EffectiveUsage returns the usage string. When SubCommands are present, +// it is auto-generated from sub-command names so metadata and behavior +// cannot drift. +func (d Definition) EffectiveUsage() string { + if len(d.SubCommands) == 0 { + return d.Usage + } + names := make([]string, 0, len(d.SubCommands)) + for _, sc := range d.SubCommands { + name := sc.Name + if sc.ArgsUsage != "" { + name += " " + sc.ArgsUsage + } + names = append(names, name) + } + return fmt.Sprintf("/%s [%s]", d.Name, strings.Join(names, "|")) +} diff --git a/pkg/commands/definition_test.go b/pkg/commands/definition_test.go new file mode 100644 index 000000000..27ad4a0a2 --- /dev/null +++ b/pkg/commands/definition_test.go @@ -0,0 +1,41 @@ +package commands + +import ( + "testing" +) + +func TestDefinition_EffectiveUsage_NoSubCommands(t *testing.T) { + d := Definition{Name: "start", Usage: "/start"} + if got := d.EffectiveUsage(); got != "/start" { + t.Fatalf("EffectiveUsage()=%q, want %q", got, "/start") + } +} + +func TestDefinition_EffectiveUsage_WithSubCommands(t *testing.T) { + d := Definition{ + Name: "show", + SubCommands: []SubCommand{ + {Name: "model"}, + {Name: "channel"}, + {Name: "agents"}, + }, + } + want := "/show [model|channel|agents]" + if got := d.EffectiveUsage(); got != want { + t.Fatalf("EffectiveUsage()=%q, want %q", got, want) + } +} + +func TestDefinition_EffectiveUsage_WithArgsUsage(t *testing.T) { + d := Definition{ + Name: "session", + SubCommands: []SubCommand{ + {Name: "list"}, + {Name: "resume", ArgsUsage: ""}, + }, + } + want := "/session [list|resume ]" + if got := d.EffectiveUsage(); got != want { + t.Fatalf("EffectiveUsage()=%q, want %q", got, want) + } +} diff --git a/pkg/commands/executor.go b/pkg/commands/executor.go new file mode 100644 index 000000000..78a50e6c2 --- /dev/null +++ b/pkg/commands/executor.go @@ -0,0 +1,89 @@ +package commands + +import ( + "context" + "fmt" +) + +type Outcome int + +const ( + // OutcomePassthrough means this input should continue through normal agent flow. + OutcomePassthrough Outcome = iota + // OutcomeHandled means a command handler executed (with or without handler error). + OutcomeHandled +) + +type ExecuteResult struct { + Outcome Outcome + Command string + Err error +} + +type Executor struct { + reg *Registry + rt *Runtime +} + +func NewExecutor(reg *Registry, rt *Runtime) *Executor { + return &Executor{reg: reg, rt: rt} +} + +// Execute implements a two-state command decision: +// 1) handled: execute command immediately; +// 2) passthrough: not a command or intentionally deferred to agent logic. +func (e *Executor) Execute(ctx context.Context, req Request) ExecuteResult { + cmdName, ok := parseCommandName(req.Text) + if !ok { + return ExecuteResult{Outcome: OutcomePassthrough} + } + + if e == nil || e.reg == nil { + return ExecuteResult{Outcome: OutcomePassthrough, Command: cmdName} + } + + def, found := e.reg.Lookup(cmdName) + if !found { + return ExecuteResult{Outcome: OutcomePassthrough, Command: cmdName} + } + + return e.executeDefinition(ctx, req, def) +} + +func (e *Executor) executeDefinition(ctx context.Context, req Request, def Definition) ExecuteResult { + // Ensure Reply is always non-nil so handlers don't need to check. + if req.Reply == nil { + req.Reply = func(string) error { return nil } + } + + // Simple command — no sub-commands + if len(def.SubCommands) == 0 { + if def.Handler == nil { + return ExecuteResult{Outcome: OutcomePassthrough, Command: def.Name} + } + err := def.Handler(ctx, req, e.rt) + return ExecuteResult{Outcome: OutcomeHandled, Command: def.Name, Err: err} + } + + // Sub-command routing + subName := nthToken(req.Text, 1) + if subName == "" { + err := req.Reply("Usage: " + def.EffectiveUsage()) + return ExecuteResult{Outcome: OutcomeHandled, Command: def.Name, Err: err} + } + + normalized := normalizeCommandName(subName) + for _, sc := range def.SubCommands { + if normalizeCommandName(sc.Name) == normalized { + if sc.Handler == nil { + return ExecuteResult{Outcome: OutcomePassthrough, Command: def.Name} + } + err := sc.Handler(ctx, req, e.rt) + return ExecuteResult{Outcome: OutcomeHandled, Command: def.Name, Err: err} + } + } + + // Unknown sub-command + err := req.Reply(fmt.Sprintf("Unknown option: %s. Usage: %s", subName, def.EffectiveUsage())) + return ExecuteResult{Outcome: OutcomeHandled, Command: def.Name, Err: err} +} diff --git a/pkg/commands/executor_test.go b/pkg/commands/executor_test.go new file mode 100644 index 000000000..09350f1b6 --- /dev/null +++ b/pkg/commands/executor_test.go @@ -0,0 +1,260 @@ +package commands + +import ( + "context" + "errors" + "strings" + "testing" +) + +func TestExecutor_RegisteredWithoutHandler_ReturnsPassthrough(t *testing.T) { + defs := []Definition{{Name: "show"}} + ex := NewExecutor(NewRegistry(defs), nil) + + res := ex.Execute(context.Background(), Request{Channel: "whatsapp", Text: "/show"}) + if res.Outcome != OutcomePassthrough { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomePassthrough) + } +} + +func TestExecutor_UnknownSlashCommand_ReturnsPassthrough(t *testing.T) { + defs := []Definition{{Name: "show"}} + ex := NewExecutor(NewRegistry(defs), nil) + + res := ex.Execute(context.Background(), Request{Channel: "telegram", Text: "/unknown"}) + if res.Outcome != OutcomePassthrough { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomePassthrough) + } +} + +func TestExecutor_SupportedCommandWithHandler_ReturnsHandled(t *testing.T) { + called := false + defs := []Definition{ + { + Name: "help", + Handler: func(context.Context, Request, *Runtime) error { + called = true + return nil + }, + }, + } + ex := NewExecutor(NewRegistry(defs), nil) + + res := ex.Execute(context.Background(), Request{Channel: "telegram", Text: "/help@my_bot"}) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if !called { + t.Fatalf("expected handler to be called") + } +} + +func TestExecutor_AliasWithoutHandler_ReturnsPassthrough(t *testing.T) { + defs := []Definition{ + { + Name: "show", + Aliases: []string{"display"}, + }, + } + ex := NewExecutor(NewRegistry(defs), nil) + + res := ex.Execute(context.Background(), Request{Channel: "whatsapp", Text: "/display"}) + if res.Outcome != OutcomePassthrough { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomePassthrough) + } + if res.Command != "show" { + t.Fatalf("command=%q, want=%q", res.Command, "show") + } +} + +func TestExecutor_AliasWithHandler_ReturnsHandled(t *testing.T) { + called := false + defs := []Definition{ + { + Name: "clear", + Aliases: []string{"reset"}, + Handler: func(context.Context, Request, *Runtime) error { + called = true + return nil + }, + }, + } + ex := NewExecutor(NewRegistry(defs), nil) + + res := ex.Execute(context.Background(), Request{Channel: "telegram", Text: "/reset"}) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if res.Command != "clear" { + t.Fatalf("command=%q, want=%q", res.Command, "clear") + } + if !called { + t.Fatalf("expected handler to be called") + } +} + +func TestExecutor_SupportedCommandWithNilHandler_ReturnsPassthrough(t *testing.T) { + defs := []Definition{ + {Name: "placeholder"}, + } + ex := NewExecutor(NewRegistry(defs), nil) + + res := ex.Execute(context.Background(), Request{Channel: "telegram", Text: "/placeholder list"}) + if res.Outcome != OutcomePassthrough { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomePassthrough) + } + if res.Command != "placeholder" { + t.Fatalf("command=%q, want=%q", res.Command, "placeholder") + } +} + +func TestExecutor_NilHandlerDoesNotMaskLaterHandler(t *testing.T) { + // With Lookup-based dispatch, the first registered definition for a name wins. + // A definition with nil Handler and no SubCommands returns Passthrough. + defs := []Definition{ + {Name: "placeholder"}, + } + ex := NewExecutor(NewRegistry(defs), nil) + + res := ex.Execute(context.Background(), Request{Channel: "telegram", Text: "/placeholder"}) + if res.Outcome != OutcomePassthrough { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomePassthrough) + } + if res.Command != "placeholder" { + t.Fatalf("command=%q, want=%q", res.Command, "placeholder") + } +} + +func TestExecutor_HandlerErrorIsPropagated(t *testing.T) { + wantErr := errors.New("handler failed") + defs := []Definition{ + { + Name: "help", + Handler: func(context.Context, Request, *Runtime) error { + return wantErr + }, + }, + } + ex := NewExecutor(NewRegistry(defs), nil) + + res := ex.Execute(context.Background(), Request{Channel: "telegram", Text: "/help"}) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if !errors.Is(res.Err, wantErr) { + t.Fatalf("err=%v, want=%v", res.Err, wantErr) + } +} + +func TestExecutor_SupportsBangPrefixAndCaseInsensitiveCommand(t *testing.T) { + called := false + defs := []Definition{ + { + Name: "help", + Handler: func(context.Context, Request, *Runtime) error { + called = true + return nil + }, + }, + } + ex := NewExecutor(NewRegistry(defs), nil) + + res := ex.Execute(context.Background(), Request{Channel: "telegram", Text: "!HELP"}) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if !called { + t.Fatalf("expected handler to be called") + } +} + +func TestExecutor_SubCommand_RoutesToCorrectHandler(t *testing.T) { + modelCalled := false + defs := []Definition{ + { + Name: "show", + SubCommands: []SubCommand{ + {Name: "model", Handler: func(_ context.Context, _ Request, _ *Runtime) error { + modelCalled = true + return nil + }}, + {Name: "channel"}, + }, + }, + } + ex := NewExecutor(NewRegistry(defs), nil) + + res := ex.Execute(context.Background(), Request{Text: "/show model"}) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if !modelCalled { + t.Fatal("model sub-command handler was not called") + } +} + +func TestExecutor_SubCommand_NoArg_RepliesUsage(t *testing.T) { + defs := []Definition{ + { + Name: "show", + SubCommands: []SubCommand{ + {Name: "model"}, + {Name: "channel"}, + }, + }, + } + ex := NewExecutor(NewRegistry(defs), nil) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/show", + Reply: func(text string) error { reply = text; return nil }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if reply != "Usage: /show [model|channel]" { + t.Fatalf("reply=%q, want usage message", reply) + } +} + +func TestExecutor_SubCommand_UnknownArg_RepliesError(t *testing.T) { + defs := []Definition{ + { + Name: "show", + SubCommands: []SubCommand{ + {Name: "model"}, + }, + }, + } + ex := NewExecutor(NewRegistry(defs), nil) + + var reply string + res := ex.Execute(context.Background(), Request{ + Text: "/show foobar", + Reply: func(text string) error { reply = text; return nil }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if !strings.Contains(reply, "foobar") { + t.Fatalf("reply=%q, should mention unknown sub-command", reply) + } +} + +func TestExecutor_SubCommand_NilHandler_ReturnsPassthrough(t *testing.T) { + defs := []Definition{ + { + Name: "show", + SubCommands: []SubCommand{ + {Name: "model"}, // nil Handler + }, + }, + } + ex := NewExecutor(NewRegistry(defs), nil) + + res := ex.Execute(context.Background(), Request{Text: "/show model"}) + if res.Outcome != OutcomePassthrough { + t.Fatalf("outcome=%v, want=%v", res.Outcome, OutcomePassthrough) + } +} diff --git a/pkg/commands/handler_agents.go b/pkg/commands/handler_agents.go new file mode 100644 index 000000000..c459516eb --- /dev/null +++ b/pkg/commands/handler_agents.go @@ -0,0 +1,21 @@ +package commands + +import ( + "context" + "fmt" + "strings" +) + +// agentsHandler returns a shared handler for both /show agents and /list agents. +func agentsHandler() Handler { + return func(_ context.Context, req Request, rt *Runtime) error { + if rt == nil || rt.ListAgentIDs == nil { + return req.Reply(unavailableMsg) + } + ids := rt.ListAgentIDs() + if len(ids) == 0 { + return req.Reply("No agents registered") + } + return req.Reply(fmt.Sprintf("Registered agents: %s", strings.Join(ids, ", "))) + } +} diff --git a/pkg/commands/registry.go b/pkg/commands/registry.go new file mode 100644 index 000000000..e17d489a6 --- /dev/null +++ b/pkg/commands/registry.go @@ -0,0 +1,55 @@ +package commands + +type Registry struct { + defs []Definition + index map[string]int +} + +// NewRegistry stores the canonical command set used by both dispatch and +// optional platform registration adapters. +func NewRegistry(defs []Definition) *Registry { + stored := make([]Definition, len(defs)) + copy(stored, defs) + + index := make(map[string]int, len(stored)*2) + for i, def := range stored { + registerCommandName(index, def.Name, i) + for _, alias := range def.Aliases { + registerCommandName(index, alias, i) + } + } + + return &Registry{defs: stored, index: index} +} + +// Definitions returns all registered command definitions. +// Command availability is global and no longer channel-scoped. +func (r *Registry) Definitions() []Definition { + out := make([]Definition, len(r.defs)) + copy(out, r.defs) + return out +} + +// Lookup returns a command definition by normalized command name or alias. +func (r *Registry) Lookup(name string) (Definition, bool) { + key := normalizeCommandName(name) + if key == "" { + return Definition{}, false + } + idx, ok := r.index[key] + if !ok { + return Definition{}, false + } + return r.defs[idx], true +} + +func registerCommandName(index map[string]int, name string, defIndex int) { + key := normalizeCommandName(name) + if key == "" { + return + } + if _, exists := index[key]; exists { + return + } + index[key] = defIndex +} diff --git a/pkg/commands/registry_test.go b/pkg/commands/registry_test.go new file mode 100644 index 000000000..bfff76b7c --- /dev/null +++ b/pkg/commands/registry_test.go @@ -0,0 +1,49 @@ +package commands + +import "testing" + +func TestRegistry_Definitions_ReturnsCopy(t *testing.T) { + defs := []Definition{ + {Name: "help", Description: "Show help"}, + {Name: "admin", Description: "Admin command"}, + } + r := NewRegistry(defs) + + got := r.Definitions() + if len(got) != 2 { + t.Fatalf("definitions len = %d, want 2", len(got)) + } + + got[0].Name = "mutated" + again := r.Definitions() + if again[0].Name != "help" { + t.Fatalf("registry should not be mutated by caller, got first name %q", again[0].Name) + } +} + +func TestRegistry_Lookup_MatchesByLowercaseNameAndAlias(t *testing.T) { + r := NewRegistry([]Definition{ + {Name: "Help", Aliases: []string{"Assist"}}, + {Name: "List"}, + }) + + def, ok := r.Lookup("help") + if !ok || def.Name != "Help" { + t.Fatalf("lookup by lowercase name failed: ok=%v def=%+v", ok, def) + } + + def, ok = r.Lookup("HELP") + if !ok || def.Name != "Help" { + t.Fatalf("lookup by uppercase name failed: ok=%v def=%+v", ok, def) + } + + def, ok = r.Lookup("assist") + if !ok || def.Name != "Help" { + t.Fatalf("lookup by lowercase alias failed: ok=%v def=%+v", ok, def) + } + + def, ok = r.Lookup("ASSIST") + if !ok || def.Name != "Help" { + t.Fatalf("lookup by uppercase alias failed: ok=%v def=%+v", ok, def) + } +} diff --git a/pkg/commands/request.go b/pkg/commands/request.go new file mode 100644 index 000000000..62ee600f2 --- /dev/null +++ b/pkg/commands/request.go @@ -0,0 +1,75 @@ +package commands + +import ( + "context" + "strings" +) + +type Handler func(ctx context.Context, req Request, rt *Runtime) error + +type Request struct { + Channel string + ChatID string + SenderID string + Text string + Reply func(text string) error +} + +const unavailableMsg = "Command unavailable in current context." + +var commandPrefixes = []string{"/", "!"} + +// parseCommandName accepts "/name", "!name", and Telegram's "/name@bot", then +// normalizes to lowercase command names. +func parseCommandName(input string) (string, bool) { + token := nthToken(input, 0) + if token == "" { + return "", false + } + + name, ok := trimCommandPrefix(token) + if !ok { + return "", false + } + if i := strings.Index(name, "@"); i >= 0 { + name = name[:i] + } + name = normalizeCommandName(name) + if name == "" { + return "", false + } + return name, true +} + +func trimCommandPrefix(token string) (string, bool) { + for _, prefix := range commandPrefixes { + if strings.HasPrefix(token, prefix) { + return strings.TrimPrefix(token, prefix), true + } + } + return "", false +} + +// HasCommandPrefix returns true if the input starts with a recognized +// command prefix (e.g. "/" or "!"). +func HasCommandPrefix(input string) bool { + token := nthToken(input, 0) + if token == "" { + return false + } + _, ok := trimCommandPrefix(token) + return ok +} + +// nthToken returns the 0-indexed token from whitespace-split input. +func nthToken(input string, n int) string { + parts := strings.Fields(strings.TrimSpace(input)) + if n >= len(parts) { + return "" + } + return parts[n] +} + +func normalizeCommandName(name string) string { + return strings.ToLower(strings.TrimSpace(name)) +} diff --git a/pkg/commands/request_test.go b/pkg/commands/request_test.go new file mode 100644 index 000000000..4389e453b --- /dev/null +++ b/pkg/commands/request_test.go @@ -0,0 +1,28 @@ +package commands + +import "testing" + +func TestHasCommandPrefix(t *testing.T) { + tests := []struct { + input string + want bool + }{ + {"/help", true}, + {"!help", true}, + {"/switch model to gpt-4", true}, + {"!switch model to gpt-4", true}, + {"hello", false}, + {"", false}, + {" ", false}, + {"hello /world", false}, + {"/", true}, + {"!", true}, + {" /help", true}, + } + for _, tt := range tests { + got := HasCommandPrefix(tt.input) + if got != tt.want { + t.Errorf("HasCommandPrefix(%q) = %v, want %v", tt.input, got, tt.want) + } + } +} diff --git a/pkg/commands/runtime.go b/pkg/commands/runtime.go new file mode 100644 index 000000000..227d495f4 --- /dev/null +++ b/pkg/commands/runtime.go @@ -0,0 +1,16 @@ +package commands + +import "github.com/sipeed/picoclaw/pkg/config" + +// Runtime provides runtime dependencies to command handlers. It is constructed +// per-request by the agent loop so that per-request state (like session scope) +// can coexist with long-lived callbacks (like GetModelInfo). +type Runtime struct { + Config *config.Config + GetModelInfo func() (name, provider string) + ListAgentIDs func() []string + ListDefinitions func() []Definition + GetEnabledChannels func() []string + SwitchModel func(value string) (oldModel string, err error) + SwitchChannel func(value string) error +} diff --git a/pkg/commands/show_list_handlers_test.go b/pkg/commands/show_list_handlers_test.go new file mode 100644 index 000000000..047708f0f --- /dev/null +++ b/pkg/commands/show_list_handlers_test.go @@ -0,0 +1,85 @@ +package commands + +import ( + "context" + "strings" + "testing" +) + +func TestShowListHandlers_ChannelPolicy(t *testing.T) { + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), nil) + + var telegramReply string + handled := ex.Execute(context.Background(), Request{ + Channel: "telegram", + Text: "/show channel", + Reply: func(text string) error { + telegramReply = text + return nil + }, + }) + if handled.Outcome != OutcomeHandled { + t.Fatalf("telegram /show outcome=%v, want=%v", handled.Outcome, OutcomeHandled) + } + if telegramReply != "Current Channel: telegram" { + t.Fatalf("telegram /show reply=%q, want=%q", telegramReply, "Current Channel: telegram") + } + + var whatsappReply string + handledWhatsApp := ex.Execute(context.Background(), Request{ + Channel: "whatsapp", + Text: "/show channel", + Reply: func(text string) error { + whatsappReply = text + return nil + }, + }) + if handledWhatsApp.Outcome != OutcomeHandled { + t.Fatalf("whatsapp /show outcome=%v, want=%v", handledWhatsApp.Outcome, OutcomeHandled) + } + if handledWhatsApp.Command != "show" { + t.Fatalf("whatsapp /show command=%q, want=%q", handledWhatsApp.Command, "show") + } + if whatsappReply != "Current Channel: whatsapp" { + t.Fatalf("whatsapp /show reply=%q, want=%q", whatsappReply, "Current Channel: whatsapp") + } + + passthrough := ex.Execute(context.Background(), Request{ + Channel: "whatsapp", + Text: "/foo", + }) + if passthrough.Outcome != OutcomePassthrough { + t.Fatalf("whatsapp /foo outcome=%v, want=%v", passthrough.Outcome, OutcomePassthrough) + } + if passthrough.Command != "foo" { + t.Fatalf("whatsapp /foo command=%q, want=%q", passthrough.Command, "foo") + } +} + +func TestShowListHandlers_ListHandledOnAllChannels(t *testing.T) { + rt := &Runtime{ + GetEnabledChannels: func() []string { + return []string{"telegram"} + }, + } + ex := NewExecutor(NewRegistry(BuiltinDefinitions()), rt) + + var reply string + res := ex.Execute(context.Background(), Request{ + Channel: "whatsapp", + Text: "/list channels", + Reply: func(text string) error { + reply = text + return nil + }, + }) + if res.Outcome != OutcomeHandled { + t.Fatalf("whatsapp /list outcome=%v, want=%v", res.Outcome, OutcomeHandled) + } + if res.Command != "list" { + t.Fatalf("whatsapp /list command=%q, want=%q", res.Command, "list") + } + if !strings.Contains(reply, "telegram") { + t.Fatalf("whatsapp /list reply=%q, expected enabled channels content", reply) + } +} diff --git a/pkg/config/config.go b/pkg/config/config.go index e8e0b3797..b34bbc294 100644 --- a/pkg/config/config.go +++ b/pkg/config/config.go @@ -167,21 +167,36 @@ type SessionConfig struct { IdentityLinks map[string][]string `json:"identity_links,omitempty"` } +// RoutingConfig controls the intelligent model routing feature. +// When enabled, each incoming message is scored against structural features +// (message length, code blocks, tool call history, conversation depth, attachments). +// Messages scoring below Threshold are sent to LightModel; all others use the +// agent's primary model. This reduces cost and latency for simple tasks without +// requiring any keyword matching — all scoring is language-agnostic. +type RoutingConfig struct { + Enabled bool `json:"enabled"` + LightModel string `json:"light_model"` // model_name from model_list to use for simple tasks + Threshold float64 `json:"threshold"` // complexity score in [0,1]; score >= threshold → primary model +} + type AgentDefaults struct { - Workspace string `json:"workspace" env:"PICOCLAW_AGENTS_DEFAULTS_WORKSPACE"` - RestrictToWorkspace bool `json:"restrict_to_workspace" env:"PICOCLAW_AGENTS_DEFAULTS_RESTRICT_TO_WORKSPACE"` - AllowReadOutsideWorkspace bool `json:"allow_read_outside_workspace" env:"PICOCLAW_AGENTS_DEFAULTS_ALLOW_READ_OUTSIDE_WORKSPACE"` - Provider string `json:"provider" env:"PICOCLAW_AGENTS_DEFAULTS_PROVIDER"` - ModelName string `json:"model_name,omitempty" env:"PICOCLAW_AGENTS_DEFAULTS_MODEL_NAME"` - Model string `json:"model" env:"PICOCLAW_AGENTS_DEFAULTS_MODEL"` // Deprecated: use model_name instead - ModelFallbacks []string `json:"model_fallbacks,omitempty"` - ImageModel string `json:"image_model,omitempty" env:"PICOCLAW_AGENTS_DEFAULTS_IMAGE_MODEL"` - ImageModelFallbacks []string `json:"image_model_fallbacks,omitempty"` - MaxTokens int `json:"max_tokens" env:"PICOCLAW_AGENTS_DEFAULTS_MAX_TOKENS"` - Temperature *float64 `json:"temperature,omitempty" env:"PICOCLAW_AGENTS_DEFAULTS_TEMPERATURE"` - MaxToolIterations int `json:"max_tool_iterations" env:"PICOCLAW_AGENTS_DEFAULTS_MAX_TOOL_ITERATIONS"` - TLDRIncludeMessage bool `json:"tldr_include_message" env:"PICOCLAW_AGENTS_DEFAULTS_TLDR_INCLUDE_MESSAGE"` - MaxMediaSize int `json:"max_media_size,omitempty" env:"PICOCLAW_AGENTS_DEFAULTS_MAX_MEDIA_SIZE"` + Workspace string `json:"workspace" env:"PICOCLAW_AGENTS_DEFAULTS_WORKSPACE"` + RestrictToWorkspace bool `json:"restrict_to_workspace" env:"PICOCLAW_AGENTS_DEFAULTS_RESTRICT_TO_WORKSPACE"` + AllowReadOutsideWorkspace bool `json:"allow_read_outside_workspace" env:"PICOCLAW_AGENTS_DEFAULTS_ALLOW_READ_OUTSIDE_WORKSPACE"` + Provider string `json:"provider" env:"PICOCLAW_AGENTS_DEFAULTS_PROVIDER"` + ModelName string `json:"model_name,omitempty" env:"PICOCLAW_AGENTS_DEFAULTS_MODEL_NAME"` + Model string `json:"model" env:"PICOCLAW_AGENTS_DEFAULTS_MODEL"` // Deprecated: use model_name instead + ModelFallbacks []string `json:"model_fallbacks,omitempty"` + ImageModel string `json:"image_model,omitempty" env:"PICOCLAW_AGENTS_DEFAULTS_IMAGE_MODEL"` + ImageModelFallbacks []string `json:"image_model_fallbacks,omitempty"` + MaxTokens int `json:"max_tokens" env:"PICOCLAW_AGENTS_DEFAULTS_MAX_TOKENS"` + Temperature *float64 `json:"temperature,omitempty" env:"PICOCLAW_AGENTS_DEFAULTS_TEMPERATURE"` + MaxToolIterations int `json:"max_tool_iterations" env:"PICOCLAW_AGENTS_DEFAULTS_MAX_TOOL_ITERATIONS"` + TLDRIncludeMessage bool `json:"tldr_include_message" env:"PICOCLAW_AGENTS_DEFAULTS_TLDR_INCLUDE_MESSAGE"` + SummarizeMessageThreshold int `json:"summarize_message_threshold" env:"PICOCLAW_AGENTS_DEFAULTS_SUMMARIZE_MESSAGE_THRESHOLD"` + SummarizeTokenPercent int `json:"summarize_token_percent" env:"PICOCLAW_AGENTS_DEFAULTS_SUMMARIZE_TOKEN_PERCENT"` + MaxMediaSize int `json:"max_media_size,omitempty" env:"PICOCLAW_AGENTS_DEFAULTS_MAX_MEDIA_SIZE"` + Routing *RoutingConfig `json:"routing,omitempty"` } const DefaultMaxMediaSize = 20 * 1024 * 1024 // 20 MB @@ -272,6 +287,7 @@ type FeishuConfig struct { type DiscordConfig struct { Enabled bool `json:"enabled" env:"PICOCLAW_CHANNELS_DISCORD_ENABLED"` Token string `json:"token" env:"PICOCLAW_CHANNELS_DISCORD_TOKEN"` + Proxy string `json:"proxy" env:"PICOCLAW_CHANNELS_DISCORD_PROXY"` AllowFrom FlexibleStringSlice `json:"allow_from" env:"PICOCLAW_CHANNELS_DISCORD_ALLOW_FROM"` MentionOnly bool `json:"mention_only" env:"PICOCLAW_CHANNELS_DISCORD_MENTION_ONLY"` GroupTrigger GroupTriggerConfig `json:"group_trigger,omitempty"` @@ -429,6 +445,7 @@ type ProvidersConfig struct { Antigravity ProviderConfig `json:"antigravity"` Qwen ProviderConfig `json:"qwen"` Mistral ProviderConfig `json:"mistral"` + Avian ProviderConfig `json:"avian"` } // IsEmpty checks if all provider configs are empty (no API keys or API bases set) @@ -452,7 +469,8 @@ func (p ProvidersConfig) IsEmpty() bool { p.GitHubCopilot.APIKey == "" && p.GitHubCopilot.APIBase == "" && p.Antigravity.APIKey == "" && p.Antigravity.APIBase == "" && p.Qwen.APIKey == "" && p.Qwen.APIBase == "" && - p.Mistral.APIKey == "" && p.Mistral.APIBase == "" + p.Mistral.APIKey == "" && p.Mistral.APIBase == "" && + p.Avian.APIKey == "" && p.Avian.APIBase == "" } // MarshalJSON implements custom JSON marshaling for ProvidersConfig @@ -503,6 +521,7 @@ type ModelConfig struct { RPM int `json:"rpm,omitempty"` // Requests per minute limit MaxTokensField string `json:"max_tokens_field,omitempty"` // Field name for max tokens (e.g., "max_completion_tokens") RequestTimeout int `json:"request_timeout,omitempty"` + ThinkingLevel string `json:"thinking_level,omitempty"` // Extended thinking: off|low|medium|high|xhigh|adaptive } // Validate checks if the ModelConfig has all required fields. @@ -521,6 +540,10 @@ type GatewayConfig struct { Port int `json:"port" env:"PICOCLAW_GATEWAY_PORT"` } +type ToolConfig struct { + Enabled bool `json:"enabled" env:"ENABLED"` +} + type BraveConfig struct { Enabled bool `json:"enabled" env:"PICOCLAW_TOOLS_WEB_BRAVE_ENABLED"` APIKey string `json:"api_key" env:"PICOCLAW_TOOLS_WEB_BRAVE_API_KEY"` @@ -545,11 +568,30 @@ type PerplexityConfig struct { MaxResults int `json:"max_results" env:"PICOCLAW_TOOLS_WEB_PERPLEXITY_MAX_RESULTS"` } +type SearXNGConfig struct { + Enabled bool `json:"enabled" env:"PICOCLAW_TOOLS_WEB_SEARXNG_ENABLED"` + BaseURL string `json:"base_url" env:"PICOCLAW_TOOLS_WEB_SEARXNG_BASE_URL"` + MaxResults int `json:"max_results" env:"PICOCLAW_TOOLS_WEB_SEARXNG_MAX_RESULTS"` +} + +type GLMSearchConfig struct { + Enabled bool `json:"enabled" env:"PICOCLAW_TOOLS_WEB_GLM_ENABLED"` + APIKey string `json:"api_key" env:"PICOCLAW_TOOLS_WEB_GLM_API_KEY"` + BaseURL string `json:"base_url" env:"PICOCLAW_TOOLS_WEB_GLM_BASE_URL"` + // SearchEngine specifies the search backend: "search_std" (default), + // "search_pro", "search_pro_sogou", or "search_pro_quark". + SearchEngine string `json:"search_engine" env:"PICOCLAW_TOOLS_WEB_GLM_SEARCH_ENGINE"` + MaxResults int `json:"max_results" env:"PICOCLAW_TOOLS_WEB_GLM_MAX_RESULTS"` +} + type WebToolsConfig struct { - Brave BraveConfig `json:"brave"` - Tavily TavilyConfig `json:"tavily"` - DuckDuckGo DuckDuckGoConfig `json:"duckduckgo"` - Perplexity PerplexityConfig `json:"perplexity"` + ToolConfig ` envPrefix:"PICOCLAW_TOOLS_WEB_"` + Brave BraveConfig ` json:"brave"` + Tavily TavilyConfig ` json:"tavily"` + DuckDuckGo DuckDuckGoConfig ` json:"duckduckgo"` + Perplexity PerplexityConfig ` json:"perplexity"` + SearXNG SearXNGConfig ` json:"searxng"` + GLMSearch GLMSearchConfig ` json:"glm_search"` // Proxy is an optional proxy URL for web tools (http/https/socks5/socks5h). // For authenticated proxies, prefer HTTP_PROXY/HTTPS_PROXY env vars instead of embedding credentials in config. Proxy string `json:"proxy,omitempty" env:"PICOCLAW_TOOLS_WEB_PROXY"` @@ -557,19 +599,29 @@ type WebToolsConfig struct { } type CronToolsConfig struct { - ExecTimeoutMinutes int `json:"exec_timeout_minutes" env:"PICOCLAW_TOOLS_CRON_EXEC_TIMEOUT_MINUTES"` // 0 means no timeout + ToolConfig ` envPrefix:"PICOCLAW_TOOLS_CRON_"` + ExecTimeoutMinutes int ` env:"PICOCLAW_TOOLS_CRON_EXEC_TIMEOUT_MINUTES" json:"exec_timeout_minutes"` // 0 means no timeout } type ExecConfig struct { - EnableDenyPatterns bool `json:"enable_deny_patterns" env:"PICOCLAW_TOOLS_EXEC_ENABLE_DENY_PATTERNS"` - CustomDenyPatterns []string `json:"custom_deny_patterns" env:"PICOCLAW_TOOLS_EXEC_CUSTOM_DENY_PATTERNS"` - CustomAllowPatterns []string `json:"custom_allow_patterns" env:"PICOCLAW_TOOLS_EXEC_CUSTOM_ALLOW_PATTERNS"` + ToolConfig ` envPrefix:"PICOCLAW_TOOLS_EXEC_"` + EnableDenyPatterns bool ` env:"PICOCLAW_TOOLS_EXEC_ENABLE_DENY_PATTERNS" json:"enable_deny_patterns"` + CustomDenyPatterns []string ` env:"PICOCLAW_TOOLS_EXEC_CUSTOM_DENY_PATTERNS" json:"custom_deny_patterns"` + CustomAllowPatterns []string ` env:"PICOCLAW_TOOLS_EXEC_CUSTOM_ALLOW_PATTERNS" json:"custom_allow_patterns"` + TimeoutSeconds int ` env:"PICOCLAW_TOOLS_EXEC_TIMEOUT_SECONDS" json:"timeout_seconds"` // 0 means use default (60s) +} + +type SkillsToolsConfig struct { + ToolConfig ` envPrefix:"PICOCLAW_TOOLS_SKILLS_"` + Registries SkillsRegistriesConfig ` json:"registries"` + MaxConcurrentSearches int ` json:"max_concurrent_searches" env:"PICOCLAW_TOOLS_SKILLS_MAX_CONCURRENT_SEARCHES"` + SearchCache SearchCacheConfig ` json:"search_cache"` } type MediaCleanupConfig struct { - Enabled bool `json:"enabled" env:"PICOCLAW_MEDIA_CLEANUP_ENABLED"` - MaxAge int `json:"max_age_minutes" env:"PICOCLAW_MEDIA_CLEANUP_MAX_AGE"` - Interval int `json:"interval_minutes" env:"PICOCLAW_MEDIA_CLEANUP_INTERVAL"` + ToolConfig ` envPrefix:"PICOCLAW_MEDIA_CLEANUP_"` + MaxAge int ` env:"PICOCLAW_MEDIA_CLEANUP_MAX_AGE" json:"max_age_minutes"` + Interval int ` env:"PICOCLAW_MEDIA_CLEANUP_INTERVAL" json:"interval_minutes"` } type ToolsConfig struct { @@ -581,12 +633,20 @@ type ToolsConfig struct { Skills SkillsToolsConfig `json:"skills"` MediaCleanup MediaCleanupConfig `json:"media_cleanup"` MCP MCPConfig `json:"mcp"` -} - -type SkillsToolsConfig struct { - Registries SkillsRegistriesConfig `json:"registries"` - MaxConcurrentSearches int `json:"max_concurrent_searches" env:"PICOCLAW_SKILLS_MAX_CONCURRENT_SEARCHES"` - SearchCache SearchCacheConfig `json:"search_cache"` + AppendFile ToolConfig `json:"append_file" envPrefix:"PICOCLAW_TOOLS_APPEND_FILE_"` + EditFile ToolConfig `json:"edit_file" envPrefix:"PICOCLAW_TOOLS_EDIT_FILE_"` + FindSkills ToolConfig `json:"find_skills" envPrefix:"PICOCLAW_TOOLS_FIND_SKILLS_"` + I2C ToolConfig `json:"i2c" envPrefix:"PICOCLAW_TOOLS_I2C_"` + InstallSkill ToolConfig `json:"install_skill" envPrefix:"PICOCLAW_TOOLS_INSTALL_SKILL_"` + ListDir ToolConfig `json:"list_dir" envPrefix:"PICOCLAW_TOOLS_LIST_DIR_"` + Message ToolConfig `json:"message" envPrefix:"PICOCLAW_TOOLS_MESSAGE_"` + ReadFile ToolConfig `json:"read_file" envPrefix:"PICOCLAW_TOOLS_READ_FILE_"` + SendFile ToolConfig `json:"send_file" envPrefix:"PICOCLAW_TOOLS_SEND_FILE_"` + Spawn ToolConfig `json:"spawn" envPrefix:"PICOCLAW_TOOLS_SPAWN_"` + SPI ToolConfig `json:"spi" envPrefix:"PICOCLAW_TOOLS_SPI_"` + Subagent ToolConfig `json:"subagent" envPrefix:"PICOCLAW_TOOLS_SUBAGENT_"` + WebFetch ToolConfig `json:"web_fetch" envPrefix:"PICOCLAW_TOOLS_WEB_FETCH_"` + WriteFile ToolConfig `json:"write_file" envPrefix:"PICOCLAW_TOOLS_WRITE_FILE_"` } type SearchCacheConfig struct { @@ -632,8 +692,7 @@ type MCPServerConfig struct { // MCPConfig defines configuration for all MCP servers type MCPConfig struct { - // Enabled globally enables/disables MCP integration - Enabled bool `json:"enabled" env:"PICOCLAW_TOOLS_MCP_ENABLED"` + ToolConfig `envPrefix:"PICOCLAW_TOOLS_MCP_"` // Servers is a map of server name to server configuration Servers map[string]MCPServerConfig `json:"servers,omitempty"` } @@ -819,3 +878,50 @@ func (c *Config) ValidateModelList() error { } return nil } + +func (t *ToolsConfig) IsToolEnabled(name string) bool { + switch name { + case "web": + return t.Web.Enabled + case "cron": + return t.Cron.Enabled + case "exec": + return t.Exec.Enabled + case "skills": + return t.Skills.Enabled + case "media_cleanup": + return t.MediaCleanup.Enabled + case "append_file": + return t.AppendFile.Enabled + case "edit_file": + return t.EditFile.Enabled + case "find_skills": + return t.FindSkills.Enabled + case "i2c": + return t.I2C.Enabled + case "install_skill": + return t.InstallSkill.Enabled + case "list_dir": + return t.ListDir.Enabled + case "message": + return t.Message.Enabled + case "read_file": + return t.ReadFile.Enabled + case "spawn": + return t.Spawn.Enabled + case "spi": + return t.SPI.Enabled + case "subagent": + return t.Subagent.Enabled + case "web_fetch": + return t.WebFetch.Enabled + case "send_file": + return t.SendFile.Enabled + case "write_file": + return t.WriteFile.Enabled + case "mcp": + return t.MCP.Enabled + default: + return true + } +} diff --git a/pkg/config/config_test.go b/pkg/config/config_test.go index 6af7c209e..10ebc7c90 100644 --- a/pkg/config/config_test.go +++ b/pkg/config/config_test.go @@ -435,6 +435,18 @@ func TestLoadConfig_WebToolsProxy(t *testing.T) { } // TestDefaultConfig_DMScope verifies the default dm_scope value +// TestDefaultConfig_SummarizationThresholds verifies summarization defaults +func TestDefaultConfig_SummarizationThresholds(t *testing.T) { + cfg := DefaultConfig() + + if cfg.Agents.Defaults.SummarizeMessageThreshold != 20 { + t.Errorf("SummarizeMessageThreshold = %d, want 20", cfg.Agents.Defaults.SummarizeMessageThreshold) + } + if cfg.Agents.Defaults.SummarizeTokenPercent != 75 { + t.Errorf("SummarizeTokenPercent = %d, want 75", cfg.Agents.Defaults.SummarizeTokenPercent) + } +} + func TestDefaultConfig_DMScope(t *testing.T) { cfg := DefaultConfig() diff --git a/pkg/config/defaults.go b/pkg/config/defaults.go index 9fc09c5f1..1902480c5 100644 --- a/pkg/config/defaults.go +++ b/pkg/config/defaults.go @@ -26,13 +26,15 @@ func DefaultConfig() *Config { return &Config{ Agents: AgentsConfig{ Defaults: AgentDefaults{ - Workspace: workspacePath, - RestrictToWorkspace: true, - Provider: "", - Model: "", - MaxTokens: 32768, - Temperature: nil, // nil means use provider default - MaxToolIterations: 50, + Workspace: workspacePath, + RestrictToWorkspace: true, + Provider: "", + Model: "", + MaxTokens: 32768, + Temperature: nil, // nil means use provider default + MaxToolIterations: 50, + SummarizeMessageThreshold: 20, + SummarizeTokenPercent: 75, }, }, Bindings: []AgentBinding{}, @@ -306,6 +308,20 @@ func DefaultConfig() *Config { APIKey: "", }, + // Avian - https://avian.io + { + ModelName: "deepseek-v3.2", + Model: "avian/deepseek/deepseek-v3.2", + APIBase: "https://api.avian.io/v1", + APIKey: "", + }, + { + ModelName: "kimi-k2.5", + Model: "avian/moonshotai/kimi-k2.5", + APIBase: "https://api.avian.io/v1", + APIKey: "", + }, + // VLLM (local) - http://localhost:8000 { ModelName: "local-model", @@ -320,11 +336,16 @@ func DefaultConfig() *Config { }, Tools: ToolsConfig{ MediaCleanup: MediaCleanupConfig{ - Enabled: true, + ToolConfig: ToolConfig{ + Enabled: true, + }, MaxAge: 30, Interval: 5, }, Web: WebToolsConfig{ + ToolConfig: ToolConfig{ + Enabled: true, + }, Proxy: "", FetchLimitBytes: 10 * 1024 * 1024, // 10MB by default Brave: BraveConfig{ @@ -341,14 +362,36 @@ func DefaultConfig() *Config { APIKey: "", MaxResults: 5, }, + SearXNG: SearXNGConfig{ + Enabled: false, + BaseURL: "", + MaxResults: 5, + }, + GLMSearch: GLMSearchConfig{ + Enabled: false, + APIKey: "", + BaseURL: "https://open.bigmodel.cn/api/paas/v4/web_search", + SearchEngine: "search_std", + MaxResults: 5, + }, }, Cron: CronToolsConfig{ + ToolConfig: ToolConfig{ + Enabled: true, + }, ExecTimeoutMinutes: 5, }, Exec: ExecConfig{ + ToolConfig: ToolConfig{ + Enabled: true, + }, EnableDenyPatterns: true, + TimeoutSeconds: 60, }, Skills: SkillsToolsConfig{ + ToolConfig: ToolConfig{ + Enabled: true, + }, Registries: SkillsRegistriesConfig{ ClawHub: ClawHubRegistryConfig{ Enabled: true, @@ -361,10 +404,54 @@ func DefaultConfig() *Config { TTLSeconds: 300, }, }, + SendFile: ToolConfig{ + Enabled: true, + }, MCP: MCPConfig{ - Enabled: false, + ToolConfig: ToolConfig{ + Enabled: false, + }, Servers: map[string]MCPServerConfig{}, }, + AppendFile: ToolConfig{ + Enabled: true, + }, + EditFile: ToolConfig{ + Enabled: true, + }, + FindSkills: ToolConfig{ + Enabled: true, + }, + I2C: ToolConfig{ + Enabled: false, // Hardware tool - Linux only + }, + InstallSkill: ToolConfig{ + Enabled: true, + }, + ListDir: ToolConfig{ + Enabled: true, + }, + Message: ToolConfig{ + Enabled: true, + }, + ReadFile: ToolConfig{ + Enabled: true, + }, + Spawn: ToolConfig{ + Enabled: true, + }, + SPI: ToolConfig{ + Enabled: false, // Hardware tool - Linux only + }, + Subagent: ToolConfig{ + Enabled: true, + }, + WebFetch: ToolConfig{ + Enabled: true, + }, + WriteFile: ToolConfig{ + Enabled: true, + }, }, Heartbeat: HeartbeatConfig{ Enabled: true, diff --git a/pkg/config/migration.go b/pkg/config/migration.go index 772f714fd..4a17dd6c9 100644 --- a/pkg/config/migration.go +++ b/pkg/config/migration.go @@ -373,6 +373,23 @@ func ConvertProvidersToModelList(cfg *Config) []ModelConfig { }, true }, }, + { + providerNames: []string{"avian"}, + protocol: "avian", + buildConfig: func(p ProvidersConfig) (ModelConfig, bool) { + if p.Avian.APIKey == "" && p.Avian.APIBase == "" { + return ModelConfig{}, false + } + return ModelConfig{ + ModelName: "avian", + Model: "avian/deepseek/deepseek-v3.2", + APIKey: p.Avian.APIKey, + APIBase: p.Avian.APIBase, + Proxy: p.Avian.Proxy, + RequestTimeout: p.Avian.RequestTimeout, + }, true + }, + }, } // Process each provider migration diff --git a/pkg/config/migration_test.go b/pkg/config/migration_test.go index e24e9fa1d..67ad73db9 100644 --- a/pkg/config/migration_test.go +++ b/pkg/config/migration_test.go @@ -160,14 +160,15 @@ func TestConvertProvidersToModelList_AllProviders(t *testing.T) { Antigravity: ProviderConfig{AuthMethod: "oauth"}, Qwen: ProviderConfig{APIKey: "key17"}, Mistral: ProviderConfig{APIKey: "key18"}, + Avian: ProviderConfig{APIKey: "key19"}, }, } result := ConvertProvidersToModelList(cfg) - // All 19 providers should be converted - if len(result) != 19 { - t.Errorf("len(result) = %d, want 19", len(result)) + // All 20 providers should be converted + if len(result) != 20 { + t.Errorf("len(result) = %d, want 20", len(result)) } } diff --git a/pkg/cron/service.go b/pkg/cron/service.go index 6962041c1..04775ac42 100644 --- a/pkg/cron/service.go +++ b/pkg/cron/service.go @@ -190,14 +190,21 @@ func (cs *CronService) executeJobByID(jobID string) { cs.mu.RUnlock() if callbackJob == nil { + log.Printf("[cron] job %s not found, skipping", jobID) return } + // Log job execution start + log.Printf("[cron] ▶ executing job '%s' (id: %s, schedule: %s, channel: %s)", + callbackJob.Name, jobID, callbackJob.Schedule.Kind, callbackJob.Payload.Channel) + var err error if cs.onJob != nil { _, err = cs.onJob(callbackJob) } + execDuration := time.Now().UnixMilli() - startTime + // Now acquire lock to update state cs.mu.Lock() defer cs.mu.Unlock() @@ -220,22 +227,35 @@ func (cs *CronService) executeJobByID(jobID string) { if err != nil { job.State.LastStatus = "error" job.State.LastError = err.Error() + log.Printf("[cron] ✗ job '%s' failed after %dms: %v", job.Name, execDuration, err) } else { job.State.LastStatus = "ok" job.State.LastError = "" } // Compute next run time + var nextRunStr string if job.Schedule.Kind == "at" { if job.DeleteAfterRun { cs.removeJobUnsafe(job.ID) + nextRunStr = "(deleted)" } else { job.Enabled = false job.State.NextRunAtMS = nil + nextRunStr = "(disabled)" } } else { nextRun := cs.computeNextRun(&job.Schedule, time.Now().UnixMilli()) job.State.NextRunAtMS = nextRun + if nextRun != nil { + nextRunStr = time.UnixMilli(*nextRun).Format("2006-01-02 15:04:05") + } else { + nextRunStr = "(none)" + } + } + + if err == nil { + log.Printf("[cron] ✓ job '%s' completed in %dms, next run: %s", job.Name, execDuration, nextRunStr) } if err := cs.saveStoreUnsafe(); err != nil { diff --git a/pkg/mcp/manager_test.go b/pkg/mcp/manager_test.go index 8ce81d09e..f353942ab 100644 --- a/pkg/mcp/manager_test.go +++ b/pkg/mcp/manager_test.go @@ -194,7 +194,9 @@ func TestLoadFromMCPConfig_EmptyWorkspaceWithRelativeEnvFile(t *testing.T) { mgr := NewManager() mcpCfg := config.MCPConfig{ - Enabled: true, + ToolConfig: config.ToolConfig{ + Enabled: true, + }, Servers: map[string]config.MCPServerConfig{ "test-server": { Enabled: true, @@ -228,12 +230,20 @@ func TestNewManager_InitialState(t *testing.T) { func TestLoadFromMCPConfig_DisabledOrEmptyServers(t *testing.T) { mgr := NewManager() - err := mgr.LoadFromMCPConfig(context.Background(), config.MCPConfig{Enabled: false}, "/tmp") + err := mgr.LoadFromMCPConfig( + context.Background(), + config.MCPConfig{ToolConfig: config.ToolConfig{Enabled: false}}, + "/tmp", + ) if err != nil { t.Fatalf("expected nil error when MCP disabled, got: %v", err) } - err = mgr.LoadFromMCPConfig(context.Background(), config.MCPConfig{Enabled: true}, "/tmp") + err = mgr.LoadFromMCPConfig( + context.Background(), + config.MCPConfig{ToolConfig: config.ToolConfig{Enabled: true}}, + "/tmp", + ) if err != nil { t.Fatalf("expected nil error when no servers configured, got: %v", err) } diff --git a/pkg/memory/jsonl.go b/pkg/memory/jsonl.go new file mode 100644 index 000000000..e12e2c5ab --- /dev/null +++ b/pkg/memory/jsonl.go @@ -0,0 +1,460 @@ +package memory + +import ( + "bufio" + "bytes" + "context" + "encoding/json" + "fmt" + "hash/fnv" + "log" + "os" + "path/filepath" + "strings" + "sync" + "time" + + "github.com/sipeed/picoclaw/pkg/fileutil" + "github.com/sipeed/picoclaw/pkg/providers" +) + +const ( + // numLockShards is the fixed number of mutexes used to serialize + // per-session access. Using a sharded array instead of a map keeps + // memory bounded regardless of how many sessions are created over + // the lifetime of the process — important for a long-running daemon. + numLockShards = 64 + + // maxLineSize is the maximum size of a single JSON line in a .jsonl + // file. Tool results (read_file, web search, etc.) can be large, so + // we set a generous limit. The scanner starts at 64 KB and grows + // only as needed up to this cap. + maxLineSize = 10 * 1024 * 1024 // 10 MB +) + +// sessionMeta holds per-session metadata stored in a .meta.json file. +type sessionMeta struct { + Key string `json:"key"` + Summary string `json:"summary"` + Skip int `json:"skip"` + Count int `json:"count"` + CreatedAt time.Time `json:"created_at"` + UpdatedAt time.Time `json:"updated_at"` +} + +// JSONLStore implements Store using append-only JSONL files. +// +// Each session is stored as two files: +// +// {sanitized_key}.jsonl — one JSON-encoded message per line, append-only +// {sanitized_key}.meta.json — session metadata (summary, logical truncation offset) +// +// Messages are never physically deleted from the JSONL file. Instead, +// TruncateHistory records a "skip" offset in the metadata file and +// GetHistory ignores lines before that offset. This keeps all writes +// append-only, which is both fast and crash-safe. +type JSONLStore struct { + dir string + locks [numLockShards]sync.Mutex +} + +// NewJSONLStore creates a new JSONL-backed store rooted at dir. +func NewJSONLStore(dir string) (*JSONLStore, error) { + err := os.MkdirAll(dir, 0o755) + if err != nil { + return nil, fmt.Errorf("memory: create directory: %w", err) + } + return &JSONLStore{dir: dir}, nil +} + +// sessionLock returns a mutex for the given session key. +// Keys are mapped to a fixed pool of shards via FNV hash, so +// memory usage is O(1) regardless of total session count. +func (s *JSONLStore) sessionLock(key string) *sync.Mutex { + h := fnv.New32a() + h.Write([]byte(key)) + return &s.locks[h.Sum32()%numLockShards] +} + +func (s *JSONLStore) jsonlPath(key string) string { + return filepath.Join(s.dir, sanitizeKey(key)+".jsonl") +} + +func (s *JSONLStore) metaPath(key string) string { + return filepath.Join(s.dir, sanitizeKey(key)+".meta.json") +} + +// sanitizeKey converts a session key to a safe filename component. +// Mirrors pkg/session.sanitizeFilename so that migration paths match. +// +// Note: this is a lossy mapping — "telegram:123" and "telegram_123" +// both produce the same filename. This is an intentional tradeoff: +// keys with colons (e.g. from channels) are by far the common case, +// and a bidirectional encoding (like URL-encoding) would complicate +// file listings and debugging. +func sanitizeKey(key string) string { + return strings.ReplaceAll(key, ":", "_") +} + +// readMeta loads the metadata file for a session. +// Returns a zero-value sessionMeta if the file does not exist. +func (s *JSONLStore) readMeta(key string) (sessionMeta, error) { + data, err := os.ReadFile(s.metaPath(key)) + if os.IsNotExist(err) { + return sessionMeta{Key: key}, nil + } + if err != nil { + return sessionMeta{}, fmt.Errorf("memory: read meta: %w", err) + } + var meta sessionMeta + err = json.Unmarshal(data, &meta) + if err != nil { + return sessionMeta{}, fmt.Errorf("memory: decode meta: %w", err) + } + return meta, nil +} + +// writeMeta atomically writes the metadata file using the project's +// standard WriteFileAtomic (temp + fsync + rename). +func (s *JSONLStore) writeMeta(key string, meta sessionMeta) error { + data, err := json.MarshalIndent(meta, "", " ") + if err != nil { + return fmt.Errorf("memory: encode meta: %w", err) + } + return fileutil.WriteFileAtomic(s.metaPath(key), data, 0o644) +} + +// readMessages reads valid JSON lines from a .jsonl file, skipping +// the first `skip` lines without unmarshaling them. This avoids the +// cost of json.Unmarshal on logically truncated messages. +// Malformed trailing lines (e.g. from a crash) are silently skipped. +func readMessages(path string, skip int) ([]providers.Message, error) { + f, err := os.Open(path) + if os.IsNotExist(err) { + return []providers.Message{}, nil + } + if err != nil { + return nil, fmt.Errorf("memory: open jsonl: %w", err) + } + defer f.Close() + + var msgs []providers.Message + scanner := bufio.NewScanner(f) + // Allow large lines for tool results (read_file, web search, etc.). + scanner.Buffer(make([]byte, 0, 64*1024), maxLineSize) + + lineNum := 0 + for scanner.Scan() { + line := scanner.Bytes() + if len(line) == 0 { + continue + } + lineNum++ + if lineNum <= skip { + continue + } + var msg providers.Message + if err := json.Unmarshal(line, &msg); err != nil { + // Corrupt line — likely a partial write from a crash. + // Log so operators know data was skipped, but don't + // fail the entire read; this is the standard JSONL + // recovery pattern. + log.Printf("memory: skipping corrupt line %d in %s: %v", + lineNum, filepath.Base(path), err) + continue + } + msgs = append(msgs, msg) + } + if scanner.Err() != nil { + return nil, fmt.Errorf("memory: scan jsonl: %w", scanner.Err()) + } + + if msgs == nil { + msgs = []providers.Message{} + } + return msgs, nil +} + +// countLines counts the total number of non-empty lines in a .jsonl file. +// Used by TruncateHistory to reconcile a stale meta.Count without +// the overhead of unmarshaling every message. +func countLines(path string) (int, error) { + f, err := os.Open(path) + if os.IsNotExist(err) { + return 0, nil + } + if err != nil { + return 0, fmt.Errorf("memory: open jsonl: %w", err) + } + defer f.Close() + + n := 0 + scanner := bufio.NewScanner(f) + scanner.Buffer(make([]byte, 0, 64*1024), maxLineSize) + for scanner.Scan() { + if len(scanner.Bytes()) > 0 { + n++ + } + } + return n, scanner.Err() +} + +func (s *JSONLStore) AddMessage( + _ context.Context, sessionKey, role, content string, +) error { + return s.addMsg(sessionKey, providers.Message{ + Role: role, + Content: content, + }) +} + +func (s *JSONLStore) AddFullMessage( + _ context.Context, sessionKey string, msg providers.Message, +) error { + return s.addMsg(sessionKey, msg) +} + +// addMsg is the shared implementation for AddMessage and AddFullMessage. +func (s *JSONLStore) addMsg(sessionKey string, msg providers.Message) error { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + // Append the message as a single JSON line. + line, err := json.Marshal(msg) + if err != nil { + return fmt.Errorf("memory: marshal message: %w", err) + } + line = append(line, '\n') + + f, err := os.OpenFile( + s.jsonlPath(sessionKey), + os.O_CREATE|os.O_WRONLY|os.O_APPEND, + 0o644, + ) + if err != nil { + return fmt.Errorf("memory: open jsonl for append: %w", err) + } + _, writeErr := f.Write(line) + if writeErr != nil { + f.Close() + return fmt.Errorf("memory: append message: %w", writeErr) + } + // Flush to physical storage before closing. This matches the + // durability guarantee of writeMeta and rewriteJSONL (which use + // WriteFileAtomic with fsync). Without Sync, a power loss could + // leave the append in the kernel page cache only — lost on reboot. + if syncErr := f.Sync(); syncErr != nil { + f.Close() + return fmt.Errorf("memory: sync jsonl: %w", syncErr) + } + if closeErr := f.Close(); closeErr != nil { + return fmt.Errorf("memory: close jsonl: %w", closeErr) + } + + // Update metadata. + meta, err := s.readMeta(sessionKey) + if err != nil { + return err + } + now := time.Now() + if meta.Count == 0 && meta.CreatedAt.IsZero() { + meta.CreatedAt = now + } + meta.Count++ + meta.UpdatedAt = now + + return s.writeMeta(sessionKey, meta) +} + +func (s *JSONLStore) GetHistory( + _ context.Context, sessionKey string, +) ([]providers.Message, error) { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + meta, err := s.readMeta(sessionKey) + if err != nil { + return nil, err + } + + // Pass meta.Skip so readMessages skips those lines without + // unmarshaling them — avoids wasted CPU on truncated messages. + msgs, err := readMessages(s.jsonlPath(sessionKey), meta.Skip) + if err != nil { + return nil, err + } + + return msgs, nil +} + +func (s *JSONLStore) GetSummary( + _ context.Context, sessionKey string, +) (string, error) { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + meta, err := s.readMeta(sessionKey) + if err != nil { + return "", err + } + return meta.Summary, nil +} + +func (s *JSONLStore) SetSummary( + _ context.Context, sessionKey, summary string, +) error { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + meta, err := s.readMeta(sessionKey) + if err != nil { + return err + } + now := time.Now() + if meta.CreatedAt.IsZero() { + meta.CreatedAt = now + } + meta.Summary = summary + meta.UpdatedAt = now + + return s.writeMeta(sessionKey, meta) +} + +func (s *JSONLStore) TruncateHistory( + _ context.Context, sessionKey string, keepLast int, +) error { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + meta, err := s.readMeta(sessionKey) + if err != nil { + return err + } + + // Always reconcile meta.Count with the actual line count on disk. + // A crash between the JSONL append and the meta update in addMsg + // leaves meta.Count stale (e.g. file has 101 lines but meta says + // 100). Counting lines is cheap — no unmarshal, just a scan — and + // TruncateHistory is not a hot path, so always re-count. + n, countErr := countLines(s.jsonlPath(sessionKey)) + if countErr != nil { + return countErr + } + meta.Count = n + + if keepLast <= 0 { + meta.Skip = meta.Count + } else { + effective := meta.Count - meta.Skip + if keepLast < effective { + meta.Skip = meta.Count - keepLast + } + } + meta.UpdatedAt = time.Now() + + return s.writeMeta(sessionKey, meta) +} + +func (s *JSONLStore) SetHistory( + _ context.Context, + sessionKey string, + history []providers.Message, +) error { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + meta, err := s.readMeta(sessionKey) + if err != nil { + return err + } + now := time.Now() + if meta.CreatedAt.IsZero() { + meta.CreatedAt = now + } + meta.Skip = 0 + meta.Count = len(history) + meta.UpdatedAt = now + + // Write meta BEFORE rewriting the JSONL file. If we crash between + // the two writes, meta has Skip=0 and the old file is still intact, + // so GetHistory reads from line 1 — returning "too many" messages + // rather than losing data. The next SetHistory call corrects this. + err = s.writeMeta(sessionKey, meta) + if err != nil { + return err + } + + return s.rewriteJSONL(sessionKey, history) +} + +// Compact physically rewrites the JSONL file, dropping all logically +// skipped lines. This reclaims disk space that accumulates after +// repeated TruncateHistory calls. +// +// It is safe to call at any time; if there is nothing to compact +// (skip == 0) the method returns immediately. +func (s *JSONLStore) Compact( + _ context.Context, sessionKey string, +) error { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + meta, err := s.readMeta(sessionKey) + if err != nil { + return err + } + if meta.Skip == 0 { + return nil + } + + // Read only the active messages, skipping truncated lines + // without unmarshaling them. + active, err := readMessages(s.jsonlPath(sessionKey), meta.Skip) + if err != nil { + return err + } + + // Write meta BEFORE rewriting the JSONL file. If the process + // crashes between the two writes, meta has Skip=0 and the old + // (uncompacted) file is still intact, so GetHistory reads from + // line 1 — returning previously-truncated messages rather than + // losing data. The next Compact or TruncateHistory corrects this. + meta.Skip = 0 + meta.Count = len(active) + meta.UpdatedAt = time.Now() + + err = s.writeMeta(sessionKey, meta) + if err != nil { + return err + } + + return s.rewriteJSONL(sessionKey, active) +} + +// rewriteJSONL atomically replaces the JSONL file with the given messages +// using the project's standard WriteFileAtomic (temp + fsync + rename). +func (s *JSONLStore) rewriteJSONL( + sessionKey string, msgs []providers.Message, +) error { + var buf bytes.Buffer + for i, msg := range msgs { + line, err := json.Marshal(msg) + if err != nil { + return fmt.Errorf("memory: marshal message %d: %w", i, err) + } + buf.Write(line) + buf.WriteByte('\n') + } + return fileutil.WriteFileAtomic(s.jsonlPath(sessionKey), buf.Bytes(), 0o644) +} + +func (s *JSONLStore) Close() error { + return nil +} diff --git a/pkg/memory/jsonl_test.go b/pkg/memory/jsonl_test.go new file mode 100644 index 000000000..356ff14ff --- /dev/null +++ b/pkg/memory/jsonl_test.go @@ -0,0 +1,835 @@ +package memory + +import ( + "context" + "os" + "path/filepath" + "sync" + "testing" + + "github.com/sipeed/picoclaw/pkg/providers" +) + +func newTestStore(t *testing.T) *JSONLStore { + t.Helper() + store, err := NewJSONLStore(t.TempDir()) + if err != nil { + t.Fatalf("NewJSONLStore: %v", err) + } + return store +} + +func TestNewJSONLStore_CreatesDirectory(t *testing.T) { + dir := filepath.Join(t.TempDir(), "nested", "sessions") + store, err := NewJSONLStore(dir) + if err != nil { + t.Fatalf("NewJSONLStore: %v", err) + } + defer store.Close() + + info, err := os.Stat(dir) + if err != nil { + t.Fatalf("Stat: %v", err) + } + if !info.IsDir() { + t.Errorf("expected directory, got file") + } +} + +func TestAddMessage_BasicRoundtrip(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + err := store.AddMessage(ctx, "s1", "user", "hello") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + err = store.AddMessage(ctx, "s1", "assistant", "hi there") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + + history, err := store.GetHistory(ctx, "s1") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 2 { + t.Fatalf("expected 2 messages, got %d", len(history)) + } + if history[0].Role != "user" || history[0].Content != "hello" { + t.Errorf("msg[0] = %+v", history[0]) + } + if history[1].Role != "assistant" || history[1].Content != "hi there" { + t.Errorf("msg[1] = %+v", history[1]) + } +} + +func TestAddMessage_AutoCreatesSession(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + // Adding a message to a non-existent session should work. + err := store.AddMessage(ctx, "new-session", "user", "first message") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + + history, err := store.GetHistory(ctx, "new-session") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1 message, got %d", len(history)) + } +} + +func TestAddFullMessage_WithToolCalls(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + msg := providers.Message{ + Role: "assistant", + Content: "Let me search that.", + ToolCalls: []providers.ToolCall{ + { + ID: "call_abc", + Type: "function", + Function: &providers.FunctionCall{ + Name: "web_search", + Arguments: `{"q":"golang jsonl"}`, + }, + }, + }, + } + + err := store.AddFullMessage(ctx, "tc", msg) + if err != nil { + t.Fatalf("AddFullMessage: %v", err) + } + + history, err := store.GetHistory(ctx, "tc") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1, got %d", len(history)) + } + if len(history[0].ToolCalls) != 1 { + t.Fatalf("expected 1 tool call, got %d", len(history[0].ToolCalls)) + } + tc := history[0].ToolCalls[0] + if tc.ID != "call_abc" { + t.Errorf("tool call ID = %q", tc.ID) + } + if tc.Function == nil || tc.Function.Name != "web_search" { + t.Errorf("tool call function = %+v", tc.Function) + } +} + +func TestAddFullMessage_ToolCallID(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + msg := providers.Message{ + Role: "tool", + Content: "search results here", + ToolCallID: "call_abc", + } + + err := store.AddFullMessage(ctx, "tr", msg) + if err != nil { + t.Fatalf("AddFullMessage: %v", err) + } + + history, err := store.GetHistory(ctx, "tr") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1, got %d", len(history)) + } + if history[0].ToolCallID != "call_abc" { + t.Errorf("ToolCallID = %q", history[0].ToolCallID) + } +} + +func TestGetHistory_EmptySession(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + history, err := store.GetHistory(ctx, "nonexistent") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if history == nil { + t.Fatal("expected non-nil empty slice") + } + if len(history) != 0 { + t.Errorf("expected 0 messages, got %d", len(history)) + } +} + +func TestGetHistory_Ordering(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 5; i++ { + err := store.AddMessage( + ctx, "order", + "user", + string(rune('a'+i)), + ) + if err != nil { + t.Fatalf("AddMessage(%d): %v", i, err) + } + } + + history, err := store.GetHistory(ctx, "order") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 5 { + t.Fatalf("expected 5, got %d", len(history)) + } + for i := 0; i < 5; i++ { + expected := string(rune('a' + i)) + if history[i].Content != expected { + t.Errorf("msg[%d].Content = %q, want %q", i, history[i].Content, expected) + } + } +} + +func TestSetSummary_GetSummary(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + // No summary yet. + summary, err := store.GetSummary(ctx, "s1") + if err != nil { + t.Fatalf("GetSummary: %v", err) + } + if summary != "" { + t.Errorf("expected empty, got %q", summary) + } + + // Set a summary. + err = store.SetSummary(ctx, "s1", "talked about Go") + if err != nil { + t.Fatalf("SetSummary: %v", err) + } + + summary, err = store.GetSummary(ctx, "s1") + if err != nil { + t.Fatalf("GetSummary: %v", err) + } + if summary != "talked about Go" { + t.Errorf("summary = %q", summary) + } + + // Update summary. + err = store.SetSummary(ctx, "s1", "updated summary") + if err != nil { + t.Fatalf("SetSummary: %v", err) + } + + summary, err = store.GetSummary(ctx, "s1") + if err != nil { + t.Fatalf("GetSummary: %v", err) + } + if summary != "updated summary" { + t.Errorf("summary = %q", summary) + } +} + +func TestTruncateHistory_KeepLast(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 10; i++ { + err := store.AddMessage( + ctx, "trunc", + "user", + string(rune('a'+i)), + ) + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + err := store.TruncateHistory(ctx, "trunc", 4) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + + history, err := store.GetHistory(ctx, "trunc") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 4 { + t.Fatalf("expected 4, got %d", len(history)) + } + // Should be the last 4: g, h, i, j + if history[0].Content != "g" { + t.Errorf("first kept = %q, want 'g'", history[0].Content) + } + if history[3].Content != "j" { + t.Errorf("last kept = %q, want 'j'", history[3].Content) + } +} + +func TestTruncateHistory_KeepZero(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 5; i++ { + err := store.AddMessage(ctx, "empty", "user", "msg") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + err := store.TruncateHistory(ctx, "empty", 0) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + + history, err := store.GetHistory(ctx, "empty") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 0 { + t.Errorf("expected 0, got %d", len(history)) + } +} + +func TestTruncateHistory_KeepMoreThanExists(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 3; i++ { + err := store.AddMessage(ctx, "few", "user", "msg") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + // Keep 100, but only 3 exist — should keep all. + err := store.TruncateHistory(ctx, "few", 100) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + + history, err := store.GetHistory(ctx, "few") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 3 { + t.Errorf("expected 3, got %d", len(history)) + } +} + +func TestSetHistory_ReplacesAll(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + // Add some initial messages. + for i := 0; i < 5; i++ { + err := store.AddMessage(ctx, "replace", "user", "old") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + // Replace with new history. + newHistory := []providers.Message{ + {Role: "user", Content: "new1"}, + {Role: "assistant", Content: "new2"}, + } + err := store.SetHistory(ctx, "replace", newHistory) + if err != nil { + t.Fatalf("SetHistory: %v", err) + } + + history, err := store.GetHistory(ctx, "replace") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 2 { + t.Fatalf("expected 2, got %d", len(history)) + } + if history[0].Content != "new1" || history[1].Content != "new2" { + t.Errorf("history = %+v", history) + } +} + +func TestSetHistory_ResetsSkip(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + // Add messages and truncate. + for i := 0; i < 10; i++ { + err := store.AddMessage(ctx, "skip-reset", "user", "old") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + err := store.TruncateHistory(ctx, "skip-reset", 3) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + + // SetHistory should reset skip to 0. + newHistory := []providers.Message{ + {Role: "user", Content: "fresh"}, + } + err = store.SetHistory(ctx, "skip-reset", newHistory) + if err != nil { + t.Fatalf("SetHistory: %v", err) + } + + history, err := store.GetHistory(ctx, "skip-reset") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1, got %d", len(history)) + } + if history[0].Content != "fresh" { + t.Errorf("content = %q", history[0].Content) + } +} + +func TestColonInKey(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + err := store.AddMessage(ctx, "telegram:123", "user", "hi") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + + history, err := store.GetHistory(ctx, "telegram:123") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1, got %d", len(history)) + } + + // Verify the file is named with underscore. + jsonlFile := filepath.Join(store.dir, "telegram_123.jsonl") + if _, statErr := os.Stat(jsonlFile); statErr != nil { + t.Errorf("expected file %s to exist: %v", jsonlFile, statErr) + } +} + +func TestCompact_RemovesSkippedMessages(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + // Write 10 messages, then truncate to keep last 3. + for i := 0; i < 10; i++ { + err := store.AddMessage(ctx, "compact", "user", string(rune('a'+i))) + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + err := store.TruncateHistory(ctx, "compact", 3) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + + // Before compact: file still has 10 lines. + allOnDisk, err := readMessages(store.jsonlPath("compact"), 0) + if err != nil { + t.Fatalf("readMessages: %v", err) + } + if len(allOnDisk) != 10 { + t.Fatalf("before compact: expected 10 on disk, got %d", len(allOnDisk)) + } + + // Compact. + err = store.Compact(ctx, "compact") + if err != nil { + t.Fatalf("Compact: %v", err) + } + + // After compact: file should have only 3 lines. + allOnDisk, err = readMessages(store.jsonlPath("compact"), 0) + if err != nil { + t.Fatalf("readMessages: %v", err) + } + if len(allOnDisk) != 3 { + t.Fatalf("after compact: expected 3 on disk, got %d", len(allOnDisk)) + } + + // GetHistory should still return the same 3 messages. + history, err := store.GetHistory(ctx, "compact") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 3 { + t.Fatalf("expected 3, got %d", len(history)) + } + if history[0].Content != "h" || history[2].Content != "j" { + t.Errorf("wrong content: %+v", history) + } +} + +func TestCompact_NoOpWhenNoSkip(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 5; i++ { + err := store.AddMessage(ctx, "noop", "user", "msg") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + // Compact without prior truncation — should be a no-op. + err := store.Compact(ctx, "noop") + if err != nil { + t.Fatalf("Compact: %v", err) + } + + history, err := store.GetHistory(ctx, "noop") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 5 { + t.Errorf("expected 5, got %d", len(history)) + } +} + +func TestCompact_ThenAppend(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 8; i++ { + err := store.AddMessage(ctx, "cap", "user", string(rune('a'+i))) + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + err := store.TruncateHistory(ctx, "cap", 2) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + err = store.Compact(ctx, "cap") + if err != nil { + t.Fatalf("Compact: %v", err) + } + + // Append after compaction should work correctly. + err = store.AddMessage(ctx, "cap", "user", "new") + if err != nil { + t.Fatalf("AddMessage after compact: %v", err) + } + + history, err := store.GetHistory(ctx, "cap") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 3 { + t.Fatalf("expected 3, got %d", len(history)) + } + // g, h (kept from truncation), new (appended after compaction). + if history[0].Content != "g" { + t.Errorf("first = %q, want 'g'", history[0].Content) + } + if history[2].Content != "new" { + t.Errorf("last = %q, want 'new'", history[2].Content) + } +} + +func TestTruncateHistory_StaleMetaCount(t *testing.T) { + // Simulates a crash between JSONL append and meta update in addMsg: + // file has N+1 lines but meta.Count is still N. TruncateHistory must + // reconcile with the real line count so that keepLast is accurate. + store := newTestStore(t) + ctx := context.Background() + + // Write 10 messages normally (meta.Count = 10). + for i := 0; i < 10; i++ { + err := store.AddMessage(ctx, "stale", "user", string(rune('a'+i))) + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + // Simulate crash: append a line to JSONL but do NOT update meta. + // This leaves meta.Count = 10 while the file has 11 lines. + jsonlPath := store.jsonlPath("stale") + f, err := os.OpenFile(jsonlPath, os.O_WRONLY|os.O_APPEND, 0o644) + if err != nil { + t.Fatalf("open for append: %v", err) + } + _, err = f.WriteString(`{"role":"user","content":"orphan"}` + "\n") + if err != nil { + t.Fatalf("write orphan: %v", err) + } + f.Close() + + // TruncateHistory(keepLast=4) should keep the last 4 of 11 lines, + // not the last 4 of 10. + err = store.TruncateHistory(ctx, "stale", 4) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + + history, err := store.GetHistory(ctx, "stale") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 4 { + t.Fatalf("expected 4, got %d", len(history)) + } + // Last 4 of [a,b,c,d,e,f,g,h,i,j,orphan] = [h,i,j,orphan] + if history[0].Content != "h" { + t.Errorf("first kept = %q, want 'h'", history[0].Content) + } + if history[3].Content != "orphan" { + t.Errorf("last kept = %q, want 'orphan'", history[3].Content) + } +} + +func TestCrashRecovery_PartialLine(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + // Write a valid message first. + err := store.AddMessage(ctx, "crash", "user", "valid") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + + // Simulate a crash by appending a partial JSON line directly. + jsonlPath := store.jsonlPath("crash") + f, err := os.OpenFile(jsonlPath, os.O_WRONLY|os.O_APPEND, 0o644) + if err != nil { + t.Fatalf("open for append: %v", err) + } + _, err = f.WriteString(`{"role":"user","content":"incomple`) + if err != nil { + t.Fatalf("write partial: %v", err) + } + f.Close() + + // GetHistory should return only the valid message. + history, err := store.GetHistory(ctx, "crash") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1 valid message, got %d", len(history)) + } + if history[0].Content != "valid" { + t.Errorf("content = %q", history[0].Content) + } +} + +func TestPersistence_AcrossInstances(t *testing.T) { + dir := t.TempDir() + ctx := context.Background() + + // Write with first instance. + store1, err := NewJSONLStore(dir) + if err != nil { + t.Fatalf("NewJSONLStore: %v", err) + } + err = store1.AddMessage(ctx, "persist", "user", "remember me") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + err = store1.SetSummary(ctx, "persist", "a test session") + if err != nil { + t.Fatalf("SetSummary: %v", err) + } + store1.Close() + + // Read with second instance. + store2, err := NewJSONLStore(dir) + if err != nil { + t.Fatalf("NewJSONLStore: %v", err) + } + defer store2.Close() + + history, err := store2.GetHistory(ctx, "persist") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 || history[0].Content != "remember me" { + t.Errorf("history = %+v", history) + } + + summary, err := store2.GetSummary(ctx, "persist") + if err != nil { + t.Fatalf("GetSummary: %v", err) + } + if summary != "a test session" { + t.Errorf("summary = %q", summary) + } +} + +func TestConcurrent_AddAndRead(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + var wg sync.WaitGroup + const goroutines = 10 + const msgsPerGoroutine = 20 + + // Concurrent writes. + for g := 0; g < goroutines; g++ { + wg.Add(1) + go func() { + defer wg.Done() + for i := 0; i < msgsPerGoroutine; i++ { + _ = store.AddMessage(ctx, "concurrent", "user", "msg") + } + }() + } + wg.Wait() + + history, err := store.GetHistory(ctx, "concurrent") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + expected := goroutines * msgsPerGoroutine + if len(history) != expected { + t.Errorf("expected %d messages, got %d", expected, len(history)) + } +} + +func TestConcurrent_SummarizeRace(t *testing.T) { + // Simulates the #704 race: one goroutine adds messages while + // another truncates + sets summary — like summarizeSession(). + store := newTestStore(t) + ctx := context.Background() + + // Seed with some messages. + for i := 0; i < 20; i++ { + err := store.AddMessage(ctx, "race", "user", "seed") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + var wg sync.WaitGroup + + // Writer goroutine (main agent loop). + wg.Add(1) + go func() { + defer wg.Done() + for i := 0; i < 50; i++ { + _ = store.AddMessage(ctx, "race", "user", "new") + } + }() + + // Summarizer goroutine (background task). + wg.Add(1) + go func() { + defer wg.Done() + for i := 0; i < 10; i++ { + _ = store.SetSummary(ctx, "race", "summary") + _ = store.TruncateHistory(ctx, "race", 5) + } + }() + + wg.Wait() + + // Verify the store is still in a consistent state. + _, err := store.GetHistory(ctx, "race") + if err != nil { + t.Fatalf("GetHistory after race: %v", err) + } + _, err = store.GetSummary(ctx, "race") + if err != nil { + t.Fatalf("GetSummary after race: %v", err) + } +} + +func TestMultipleSessions_Isolation(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + err := store.AddMessage(ctx, "s1", "user", "msg for s1") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + err = store.AddMessage(ctx, "s2", "user", "msg for s2") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + + h1, err := store.GetHistory(ctx, "s1") + if err != nil { + t.Fatalf("GetHistory s1: %v", err) + } + h2, err := store.GetHistory(ctx, "s2") + if err != nil { + t.Fatalf("GetHistory s2: %v", err) + } + + if len(h1) != 1 || h1[0].Content != "msg for s1" { + t.Errorf("s1 history = %+v", h1) + } + if len(h2) != 1 || h2[0].Content != "msg for s2" { + t.Errorf("s2 history = %+v", h2) + } +} + +func BenchmarkAddMessage(b *testing.B) { + dir := b.TempDir() + store, err := NewJSONLStore(dir) + if err != nil { + b.Fatalf("NewJSONLStore: %v", err) + } + defer store.Close() + ctx := context.Background() + + b.ResetTimer() + for i := 0; i < b.N; i++ { + _ = store.AddMessage(ctx, "bench", "user", "benchmark message content") + } +} + +func BenchmarkGetHistory_100(b *testing.B) { + dir := b.TempDir() + store, err := NewJSONLStore(dir) + if err != nil { + b.Fatalf("NewJSONLStore: %v", err) + } + defer store.Close() + ctx := context.Background() + + for i := 0; i < 100; i++ { + _ = store.AddMessage(ctx, "bench", "user", "message content") + } + + b.ResetTimer() + for i := 0; i < b.N; i++ { + _, _ = store.GetHistory(ctx, "bench") + } +} + +func BenchmarkGetHistory_1000(b *testing.B) { + dir := b.TempDir() + store, err := NewJSONLStore(dir) + if err != nil { + b.Fatalf("NewJSONLStore: %v", err) + } + defer store.Close() + ctx := context.Background() + + for i := 0; i < 1000; i++ { + _ = store.AddMessage(ctx, "bench", "user", "message content") + } + + b.ResetTimer() + for i := 0; i < b.N; i++ { + _, _ = store.GetHistory(ctx, "bench") + } +} diff --git a/pkg/memory/migration.go b/pkg/memory/migration.go new file mode 100644 index 000000000..c9d5176ab --- /dev/null +++ b/pkg/memory/migration.go @@ -0,0 +1,108 @@ +package memory + +import ( + "context" + "encoding/json" + "fmt" + "log" + "os" + "path/filepath" + "strings" + "time" + + "github.com/sipeed/picoclaw/pkg/providers" +) + +// jsonSession mirrors pkg/session.Session for migration purposes. +type jsonSession struct { + Key string `json:"key"` + Messages []providers.Message `json:"messages"` + Summary string `json:"summary,omitempty"` + Created time.Time `json:"created"` + Updated time.Time `json:"updated"` +} + +// MigrateFromJSON reads legacy sessions/*.json files from sessionsDir, +// writes them into the Store, and renames each migrated file to +// .json.migrated as a backup. Returns the number of sessions migrated. +// +// Files that fail to parse are logged and skipped. Already-migrated +// files (.json.migrated) are ignored, making the function idempotent. +func MigrateFromJSON( + ctx context.Context, sessionsDir string, store Store, +) (int, error) { + entries, err := os.ReadDir(sessionsDir) + if os.IsNotExist(err) { + return 0, nil + } + if err != nil { + return 0, fmt.Errorf("memory: read sessions dir: %w", err) + } + + migrated := 0 + for _, entry := range entries { + if entry.IsDir() { + continue + } + name := entry.Name() + if !strings.HasSuffix(name, ".json") { + continue + } + // Skip already-migrated files. + if strings.HasSuffix(name, ".migrated") { + continue + } + + srcPath := filepath.Join(sessionsDir, name) + + data, readErr := os.ReadFile(srcPath) + if readErr != nil { + log.Printf("memory: migrate: skip %s: %v", name, readErr) + continue + } + + var sess jsonSession + if parseErr := json.Unmarshal(data, &sess); parseErr != nil { + log.Printf("memory: migrate: skip %s: %v", name, parseErr) + continue + } + + // Use the key from the JSON content, not the filename. + // Filenames are sanitized (":" → "_") but keys are not. + key := sess.Key + if key == "" { + key = strings.TrimSuffix(name, ".json") + } + + // Use SetHistory (atomic replace) instead of per-message + // AddFullMessage. This makes migration idempotent: if the + // process crashes after writing messages but before the + // rename below, a retry replaces the partial data cleanly + // instead of duplicating messages. + if setErr := store.SetHistory(ctx, key, sess.Messages); setErr != nil { + return migrated, fmt.Errorf( + "memory: migrate %s: set history: %w", + name, setErr, + ) + } + + if sess.Summary != "" { + if sumErr := store.SetSummary(ctx, key, sess.Summary); sumErr != nil { + return migrated, fmt.Errorf( + "memory: migrate %s: set summary: %w", + name, sumErr, + ) + } + } + + // Rename to .migrated as backup (not delete). + renameErr := os.Rename(srcPath, srcPath+".migrated") + if renameErr != nil { + log.Printf("memory: migrate: rename %s: %v", name, renameErr) + } + + migrated++ + } + + return migrated, nil +} diff --git a/pkg/memory/migration_test.go b/pkg/memory/migration_test.go new file mode 100644 index 000000000..3170758b7 --- /dev/null +++ b/pkg/memory/migration_test.go @@ -0,0 +1,384 @@ +package memory + +import ( + "context" + "encoding/json" + "os" + "path/filepath" + "testing" + "time" + + "github.com/sipeed/picoclaw/pkg/providers" +) + +func writeJSONSession( + t *testing.T, dir string, filename string, sess jsonSession, +) { + t.Helper() + data, err := json.MarshalIndent(sess, "", " ") + if err != nil { + t.Fatalf("marshal session: %v", err) + } + err = os.WriteFile(filepath.Join(dir, filename), data, 0o644) + if err != nil { + t.Fatalf("write session file: %v", err) + } +} + +func TestMigrateFromJSON_Basic(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + writeJSONSession(t, sessionsDir, "test.json", jsonSession{ + Key: "test", + Messages: []providers.Message{ + {Role: "user", Content: "hello"}, + {Role: "assistant", Content: "hi"}, + }, + Summary: "A greeting.", + Created: time.Now(), + Updated: time.Now(), + }) + + count, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + if count != 1 { + t.Errorf("expected 1 migrated, got %d", count) + } + + history, err := store.GetHistory(ctx, "test") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 2 { + t.Fatalf("expected 2 messages, got %d", len(history)) + } + if history[0].Content != "hello" || history[1].Content != "hi" { + t.Errorf("unexpected messages: %+v", history) + } + + summary, err := store.GetSummary(ctx, "test") + if err != nil { + t.Fatalf("GetSummary: %v", err) + } + if summary != "A greeting." { + t.Errorf("summary = %q", summary) + } +} + +func TestMigrateFromJSON_WithToolCalls(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + writeJSONSession(t, sessionsDir, "tools.json", jsonSession{ + Key: "tools", + Messages: []providers.Message{ + { + Role: "assistant", + Content: "Searching...", + ToolCalls: []providers.ToolCall{ + { + ID: "call_1", + Type: "function", + Function: &providers.FunctionCall{ + Name: "web_search", + Arguments: `{"q":"test"}`, + }, + }, + }, + }, + { + Role: "tool", + Content: "result", + ToolCallID: "call_1", + }, + }, + Created: time.Now(), + Updated: time.Now(), + }) + + count, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + if count != 1 { + t.Errorf("expected 1, got %d", count) + } + + history, err := store.GetHistory(ctx, "tools") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 2 { + t.Fatalf("expected 2 messages, got %d", len(history)) + } + if len(history[0].ToolCalls) != 1 { + t.Fatalf("expected 1 tool call, got %d", len(history[0].ToolCalls)) + } + if history[0].ToolCalls[0].Function.Name != "web_search" { + t.Errorf("function = %q", history[0].ToolCalls[0].Function.Name) + } + if history[1].ToolCallID != "call_1" { + t.Errorf("ToolCallID = %q", history[1].ToolCallID) + } +} + +func TestMigrateFromJSON_MultipleFiles(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 3; i++ { + key := string(rune('a' + i)) + writeJSONSession(t, sessionsDir, key+".json", jsonSession{ + Key: key, + Messages: []providers.Message{{Role: "user", Content: "msg " + key}}, + Created: time.Now(), + Updated: time.Now(), + }) + } + + count, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + if count != 3 { + t.Errorf("expected 3, got %d", count) + } + + for i := 0; i < 3; i++ { + key := string(rune('a' + i)) + history, histErr := store.GetHistory(ctx, key) + if histErr != nil { + t.Fatalf("GetHistory(%q): %v", key, histErr) + } + if len(history) != 1 { + t.Errorf("session %q: expected 1 msg, got %d", key, len(history)) + } + } +} + +func TestMigrateFromJSON_InvalidJSON(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + // One valid, one invalid. + writeJSONSession(t, sessionsDir, "good.json", jsonSession{ + Key: "good", + Messages: []providers.Message{{Role: "user", Content: "ok"}}, + Created: time.Now(), + Updated: time.Now(), + }) + err := os.WriteFile( + filepath.Join(sessionsDir, "bad.json"), + []byte("{invalid json"), + 0o644, + ) + if err != nil { + t.Fatalf("write bad file: %v", err) + } + + count, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + if count != 1 { + t.Errorf("expected 1 (bad file skipped), got %d", count) + } + + history, err := store.GetHistory(ctx, "good") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Errorf("expected 1 message, got %d", len(history)) + } +} + +func TestMigrateFromJSON_RenamesFiles(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + writeJSONSession(t, sessionsDir, "rename.json", jsonSession{ + Key: "rename", + Messages: []providers.Message{{Role: "user", Content: "hi"}}, + Created: time.Now(), + Updated: time.Now(), + }) + + _, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + + // Original .json should not exist. + _, statErr := os.Stat(filepath.Join(sessionsDir, "rename.json")) + if !os.IsNotExist(statErr) { + t.Error("rename.json should have been renamed") + } + // .json.migrated should exist. + _, statErr = os.Stat( + filepath.Join(sessionsDir, "rename.json.migrated"), + ) + if statErr != nil { + t.Errorf("rename.json.migrated should exist: %v", statErr) + } +} + +func TestMigrateFromJSON_Idempotent(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + writeJSONSession(t, sessionsDir, "idem.json", jsonSession{ + Key: "idem", + Messages: []providers.Message{{Role: "user", Content: "once"}}, + Created: time.Now(), + Updated: time.Now(), + }) + + count1, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("first migration: %v", err) + } + if count1 != 1 { + t.Errorf("first run: expected 1, got %d", count1) + } + + // Second run should find only .migrated files, skip them. + count2, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("second migration: %v", err) + } + if count2 != 0 { + t.Errorf("second run: expected 0, got %d", count2) + } + + history, err := store.GetHistory(ctx, "idem") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Errorf("expected 1 message, got %d", len(history)) + } +} + +func TestMigrateFromJSON_ColonInKey(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + // File is named telegram_123 (sanitized), but the key inside is telegram:123. + writeJSONSession(t, sessionsDir, "telegram_123.json", jsonSession{ + Key: "telegram:123", + Messages: []providers.Message{{Role: "user", Content: "from telegram"}}, + Created: time.Now(), + Updated: time.Now(), + }) + + count, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + if count != 1 { + t.Errorf("expected 1, got %d", count) + } + + // Accessible via the original key "telegram:123". + history, err := store.GetHistory(ctx, "telegram:123") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1 message, got %d", len(history)) + } + if history[0].Content != "from telegram" { + t.Errorf("content = %q", history[0].Content) + } + + // In the file-based store, "telegram:123" and "telegram_123" both + // sanitize to the same filename, so they share storage. This is + // expected — the colon-to-underscore mapping is a one-way function. + history2, err := store.GetHistory(ctx, "telegram_123") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history2) != 1 { + t.Errorf("expected 1 (same file), got %d", len(history2)) + } +} + +func TestMigrateFromJSON_RetryAfterCrash(t *testing.T) { + // Simulates a crash during migration: first run writes messages + // but doesn't rename the .json file. Second run must replace + // (not duplicate) the messages thanks to SetHistory semantics. + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + writeJSONSession(t, sessionsDir, "retry.json", jsonSession{ + Key: "retry", + Messages: []providers.Message{ + {Role: "user", Content: "one"}, + {Role: "assistant", Content: "two"}, + }, + Created: time.Now(), + Updated: time.Now(), + }) + + // First migration succeeds — writes messages and renames file. + count, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("first migration: %v", err) + } + if count != 1 { + t.Fatalf("expected 1, got %d", count) + } + + // Simulate "crash before rename": restore the .json file. + src := filepath.Join(sessionsDir, "retry.json.migrated") + dst := filepath.Join(sessionsDir, "retry.json") + if renameErr := os.Rename(src, dst); renameErr != nil { + t.Fatalf("restore .json: %v", renameErr) + } + + // Second migration should re-import without duplicating messages. + count, err = MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("second migration: %v", err) + } + if count != 1 { + t.Fatalf("expected 1, got %d", count) + } + + history, err := store.GetHistory(ctx, "retry") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + // Must be exactly 2 messages (not 4 from duplication). + if len(history) != 2 { + t.Fatalf("expected 2 messages (no duplicates), got %d", len(history)) + } + if history[0].Content != "one" || history[1].Content != "two" { + t.Errorf("unexpected messages: %+v", history) + } +} + +func TestMigrateFromJSON_NonexistentDir(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + count, err := MigrateFromJSON(ctx, "/nonexistent/path", store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + if count != 0 { + t.Errorf("expected 0, got %d", count) + } +} diff --git a/pkg/memory/store.go b/pkg/memory/store.go new file mode 100644 index 000000000..b6e11707d --- /dev/null +++ b/pkg/memory/store.go @@ -0,0 +1,42 @@ +package memory + +import ( + "context" + + "github.com/sipeed/picoclaw/pkg/providers" +) + +// Store defines an interface for persistent session storage. +// Each method is an atomic operation — there is no separate Save() call. +type Store interface { + // AddMessage appends a simple text message to a session. + AddMessage(ctx context.Context, sessionKey, role, content string) error + + // AddFullMessage appends a complete message (with tool calls, etc.) to a session. + AddFullMessage(ctx context.Context, sessionKey string, msg providers.Message) error + + // GetHistory returns all messages for a session in insertion order. + // Returns an empty slice (not nil) if the session does not exist. + GetHistory(ctx context.Context, sessionKey string) ([]providers.Message, error) + + // GetSummary returns the conversation summary for a session. + // Returns an empty string if no summary exists. + GetSummary(ctx context.Context, sessionKey string) (string, error) + + // SetSummary updates the conversation summary for a session. + SetSummary(ctx context.Context, sessionKey, summary string) error + + // TruncateHistory removes all but the last keepLast messages from a session. + // If keepLast <= 0, all messages are removed. + TruncateHistory(ctx context.Context, sessionKey string, keepLast int) error + + // SetHistory replaces all messages in a session with the provided history. + SetHistory(ctx context.Context, sessionKey string, history []providers.Message) error + + // Compact reclaims storage by physically removing logically truncated + // data. Backends that do not accumulate dead data may return nil. + Compact(ctx context.Context, sessionKey string) error + + // Close releases any resources held by the store. + Close() error +} diff --git a/pkg/providers/anthropic/provider.go b/pkg/providers/anthropic/provider.go index 1bb15f771..242ded175 100644 --- a/pkg/providers/anthropic/provider.go +++ b/pkg/providers/anthropic/provider.go @@ -23,7 +23,10 @@ type ( ToolFunctionDefinition = protocoltypes.ToolFunctionDefinition ) -const defaultBaseURL = "https://api.anthropic.com" +const ( + defaultBaseURL = "https://api.anthropic.com" + anthropicBetaHeader = "oauth-2025-04-20" +) type Provider struct { client *anthropic.Client @@ -31,6 +34,9 @@ type Provider struct { baseURL string } +// SupportsThinking implements providers.ThinkingCapable. +func (p *Provider) SupportsThinking() bool { return true } + func NewProvider(token string) *Provider { return NewProviderWithBaseURL(token, "") } @@ -77,7 +83,10 @@ func (p *Provider) Chat( if err != nil { return nil, fmt.Errorf("refreshing token: %w", err) } - opts = append(opts, option.WithAuthToken(tok)) + opts = append(opts, + option.WithAuthToken(tok), + option.WithHeader("anthropic-beta", anthropicBetaHeader), + ) } params, err := buildParams(messages, tools, model, options) @@ -85,6 +94,11 @@ func (p *Provider) Chat( return nil, err } + // OAuth/setup-tokens require streaming; API keys use non-streaming. + if p.tokenSource != nil { + return p.chatStreaming(ctx, params, opts) + } + resp, err := p.client.Messages.New(ctx, params, opts...) if err != nil { return nil, fmt.Errorf("claude API call: %w", err) @@ -93,6 +107,28 @@ func (p *Provider) Chat( return parseResponse(resp), nil } +func (p *Provider) chatStreaming( + ctx context.Context, + params anthropic.MessageNewParams, + opts []option.RequestOption, +) (*LLMResponse, error) { + stream := p.client.Messages.NewStreaming(ctx, params, opts...) + defer stream.Close() + + var msg anthropic.Message + for stream.Next() { + event := stream.Current() + if err := msg.Accumulate(event); err != nil { + return nil, fmt.Errorf("claude streaming accumulate: %w", err) + } + } + if err := stream.Err(); err != nil { + return nil, fmt.Errorf("claude API call: %w", err) + } + + return parseResponse(&msg), nil +} + func (p *Provider) GetDefaultModel() string { return "claude-sonnet-4.6" } @@ -144,7 +180,16 @@ func buildParams( blocks = append(blocks, anthropic.NewTextBlock(msg.Content)) } for _, tc := range msg.ToolCalls { - blocks = append(blocks, anthropic.NewToolUseBlock(tc.ID, tc.Arguments, tc.Name)) + args := tc.Arguments + if args == nil && tc.Function != nil && tc.Function.Arguments != "" { + if err := json.Unmarshal([]byte(tc.Function.Arguments), &args); err != nil { + args = map[string]any{} + } + } + if args == nil { + args = map[string]any{} + } + blocks = append(blocks, anthropic.NewToolUseBlock(tc.ID, args, tc.Name)) } anthropicMessages = append(anthropicMessages, anthropic.NewAssistantMessage(blocks...)) } else { @@ -164,8 +209,12 @@ func buildParams( maxTokens = int64(mt) } + // Normalize model ID: Anthropic API uses hyphens (claude-sonnet-4-6), + // but config may use dots (claude-sonnet-4.6). + apiModel := strings.ReplaceAll(model, ".", "-") + params := anthropic.MessageNewParams{ - Model: anthropic.Model(model), + Model: anthropic.Model(apiModel), Messages: anthropicMessages, MaxTokens: maxTokens, } @@ -182,9 +231,80 @@ func buildParams( params.Tools = translateTools(tools) } + // Extended Thinking / Adaptive Thinking + // The thinking_level value directly determines the API parameter format: + // "adaptive" → {thinking: {type: "adaptive"}} + output_config.effort + // "low/medium/high/xhigh" → {thinking: {type: "enabled", budget_tokens: N}} + if level, ok := options["thinking_level"].(string); ok && level != "" && level != "off" { + applyThinkingConfig(¶ms, level) + } + return params, nil } +// applyThinkingConfig sets thinking parameters based on the level value. +// "adaptive" uses the adaptive thinking API (Claude 4.6+). +// All other levels use budget_tokens which is universally supported. +// +// Anthropic API constraint: temperature must not be set when thinking is enabled. +// budget_tokens must be strictly less than max_tokens. +func applyThinkingConfig(params *anthropic.MessageNewParams, level string) { + // Anthropic API rejects requests with temperature set alongside thinking. + // Reset to zero value (omitted from JSON serialization). + if params.Temperature.Valid() { + log.Printf("anthropic: temperature cleared because thinking is enabled (level=%s)", level) + } + params.Temperature = anthropic.MessageNewParams{}.Temperature + + if level == "adaptive" { + adaptive := anthropic.NewThinkingConfigAdaptiveParam() + params.Thinking = anthropic.ThinkingConfigParamUnion{OfAdaptive: &adaptive} + params.OutputConfig = anthropic.OutputConfigParam{ + Effort: anthropic.OutputConfigEffortHigh, + } + return + } + + budget := int64(levelToBudget(level)) + if budget <= 0 { + return + } + + // budget_tokens must be < max_tokens; clamp to respect user's max_tokens setting. + if budget >= params.MaxTokens { + log.Printf("anthropic: budget_tokens (%d) clamped to %d (max_tokens-1)", budget, params.MaxTokens-1) + budget = params.MaxTokens - 1 + } else if budget > params.MaxTokens*80/100 { + log.Printf("anthropic: thinking budget (%d) exceeds 80%% of max_tokens (%d), output may be truncated", + budget, params.MaxTokens) + } + params.Thinking = anthropic.ThinkingConfigParamOfEnabled(budget) +} + +// levelToBudget maps a thinking level to budget_tokens. +// Values are based on Anthropic's recommendations and community best practices: +// +// low = 4,096 — simple reasoning, quick debugging (Claude Code "think") +// medium = 16,384 — Anthropic recommended sweet spot for most tasks +// high = 32,000 — complex architecture, deep analysis (diminishing returns above this) +// xhigh = 64,000 — extreme reasoning, research problems, benchmarks +// +// Note: For Claude 4.6+, prefer adaptive thinking over manual budget_tokens. +func levelToBudget(level string) int { + switch level { + case "low": + return 4096 + case "medium": + return 16384 + case "high": + return 32000 + case "xhigh": + return 64000 + default: + return 0 + } +} + func translateTools(tools []ToolDefinition) []anthropic.ToolUnionParam { result := make([]anthropic.ToolUnionParam, 0, len(tools)) for _, t := range tools { @@ -213,10 +333,14 @@ func translateTools(tools []ToolDefinition) []anthropic.ToolUnionParam { func parseResponse(resp *anthropic.Message) *LLMResponse { var content strings.Builder + var reasoning strings.Builder var toolCalls []ToolCall for _, block := range resp.Content { switch block.Type { + case "thinking": + tb := block.AsThinking() + reasoning.WriteString(tb.Thinking) case "text": tb := block.AsText() content.WriteString(tb.Text) @@ -247,6 +371,7 @@ func parseResponse(resp *anthropic.Message) *LLMResponse { return &LLMResponse{ Content: content.String(), + Reasoning: reasoning.String(), ToolCalls: toolCalls, FinishReason: finishReason, Usage: &UsageInfo{ diff --git a/pkg/providers/anthropic/provider_test.go b/pkg/providers/anthropic/provider_test.go index 3d21c1d0b..b1aed17b5 100644 --- a/pkg/providers/anthropic/provider_test.go +++ b/pkg/providers/anthropic/provider_test.go @@ -21,8 +21,8 @@ func TestBuildParams_BasicMessage(t *testing.T) { if err != nil { t.Fatalf("buildParams() error: %v", err) } - if string(params.Model) != "claude-sonnet-4.6" { - t.Errorf("Model = %q, want %q", params.Model, "claude-sonnet-4.6") + if string(params.Model) != "claude-sonnet-4-6" { + t.Errorf("Model = %q, want %q", params.Model, "claude-sonnet-4-6") } if params.MaxTokens != 1024 { t.Errorf("MaxTokens = %d, want 1024", params.MaxTokens) @@ -262,6 +262,65 @@ func TestProvider_ChatUsesTokenSource(t *testing.T) { } } +func TestProvider_ChatStreamingRoundTrip(t *testing.T) { + server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + if r.URL.Path != "/v1/messages" { + http.Error(w, "not found", http.StatusNotFound) + return + } + if got := r.Header.Get("Authorization"); got != "Bearer refreshed-token" { + t.Errorf("Authorization = %q, want %q", got, "Bearer refreshed-token") + } + if got := r.Header.Get("Anthropic-Beta"); got != anthropicBetaHeader { + t.Errorf("Anthropic-Beta = %q, want %q", got, anthropicBetaHeader) + } + + w.Header().Set("Content-Type", "text/event-stream") + flusher, _ := w.(http.Flusher) + + events := []string{ + "event: message_start\ndata: {\"type\":\"message_start\",\"message\":{\"id\":\"msg_stream\",\"type\":\"message\",\"role\":\"assistant\",\"content\":[],\"model\":\"claude-sonnet-4-6\",\"stop_reason\":null,\"usage\":{\"input_tokens\":12,\"output_tokens\":0}}}\n\n", + "event: content_block_start\ndata: {\"type\":\"content_block_start\",\"index\":0,\"content_block\":{\"type\":\"text\",\"text\":\"\"}}\n\n", + "event: content_block_delta\ndata: {\"type\":\"content_block_delta\",\"index\":0,\"delta\":{\"type\":\"text_delta\",\"text\":\"Hello\"}}\n\n", + "event: content_block_delta\ndata: {\"type\":\"content_block_delta\",\"index\":0,\"delta\":{\"type\":\"text_delta\",\"text\":\" world\"}}\n\n", + "event: content_block_stop\ndata: {\"type\":\"content_block_stop\",\"index\":0}\n\n", + "event: message_delta\ndata: {\"type\":\"message_delta\",\"delta\":{\"stop_reason\":\"end_turn\"},\"usage\":{\"output_tokens\":5}}\n\n", + "event: message_stop\ndata: {\"type\":\"message_stop\"}\n\n", + } + for _, e := range events { + w.Write([]byte(e)) + if flusher != nil { + flusher.Flush() + } + } + })) + defer server.Close() + + p := NewProviderWithTokenSourceAndBaseURL("stale-token", func() (string, error) { + return "refreshed-token", nil + }, server.URL) + + resp, err := p.Chat( + t.Context(), + []Message{{Role: "user", Content: "Hello"}}, + nil, + "claude-sonnet-4.6", + map[string]any{}, + ) + if err != nil { + t.Fatalf("Chat() error: %v", err) + } + if resp.Content != "Hello world" { + t.Errorf("Content = %q, want %q", resp.Content, "Hello world") + } + if resp.FinishReason != "stop" { + t.Errorf("FinishReason = %q, want %q", resp.FinishReason, "stop") + } + if resp.Usage.CompletionTokens != 5 { + t.Errorf("CompletionTokens = %d, want 5", resp.Usage.CompletionTokens) + } +} + func createAnthropicTestClient(baseURL, token string) *anthropic.Client { c := anthropic.NewClient( anthropicoption.WithAuthToken(token), diff --git a/pkg/providers/anthropic/thinking_test.go b/pkg/providers/anthropic/thinking_test.go new file mode 100644 index 000000000..e69a3869e --- /dev/null +++ b/pkg/providers/anthropic/thinking_test.go @@ -0,0 +1,212 @@ +package anthropicprovider + +import ( + "encoding/json" + "testing" + + "github.com/anthropics/anthropic-sdk-go" +) + +func TestApplyThinkingConfig_Adaptive(t *testing.T) { + params := anthropic.MessageNewParams{ + MaxTokens: 16000, + Temperature: anthropic.Float(0.7), + } + applyThinkingConfig(¶ms, "adaptive") + + if params.Thinking.OfAdaptive == nil { + t.Fatal("expected adaptive thinking") + } + if params.Thinking.OfEnabled != nil { + t.Error("should not set enabled thinking in adaptive mode") + } + if params.OutputConfig.Effort != anthropic.OutputConfigEffortHigh { + t.Errorf("effort = %q, want %q", params.OutputConfig.Effort, anthropic.OutputConfigEffortHigh) + } + if params.Temperature.Valid() { + t.Error("temperature should be cleared when thinking is enabled") + } +} + +func TestApplyThinkingConfig_BudgetLevels(t *testing.T) { + tests := []struct { + level string + wantBudget int64 + }{ + {"low", 4096}, + {"medium", 16384}, + {"high", 32000}, + {"xhigh", 64000}, + } + + for _, tt := range tests { + t.Run(tt.level, func(t *testing.T) { + params := anthropic.MessageNewParams{ + MaxTokens: 200000, + Temperature: anthropic.Float(0.5), + } + applyThinkingConfig(¶ms, tt.level) + + if params.Thinking.OfEnabled == nil { + t.Fatal("expected enabled thinking") + } + if params.Thinking.OfAdaptive != nil { + t.Error("should not set adaptive thinking") + } + if params.Thinking.OfEnabled.BudgetTokens != tt.wantBudget { + t.Errorf("budget_tokens = %d, want %d", params.Thinking.OfEnabled.BudgetTokens, tt.wantBudget) + } + if params.OutputConfig.Effort != "" { + t.Errorf("effort = %q, want empty", params.OutputConfig.Effort) + } + if params.Temperature.Valid() { + t.Error("temperature should be cleared when thinking is enabled") + } + }) + } +} + +func TestApplyThinkingConfig_BudgetClamp(t *testing.T) { + // budget_tokens must be < max_tokens; clamp budget down to respect user's max_tokens. + params := anthropic.MessageNewParams{MaxTokens: 4096} + applyThinkingConfig(¶ms, "high") // budget=32000 > maxTokens=4096 + + if params.Thinking.OfEnabled == nil { + t.Fatal("expected enabled thinking") + } + if params.Thinking.OfEnabled.BudgetTokens != 4095 { + t.Errorf("budget_tokens = %d, want 4095 (maxTokens-1)", params.Thinking.OfEnabled.BudgetTokens) + } + if params.MaxTokens != 4096 { + t.Errorf("max_tokens should not be modified, got %d", params.MaxTokens) + } +} + +func TestApplyThinkingConfig_UnknownLevel(t *testing.T) { + params := anthropic.MessageNewParams{MaxTokens: 16000} + applyThinkingConfig(¶ms, "unknown") + + if params.Thinking.OfEnabled != nil { + t.Error("should not set enabled thinking for unknown level") + } + if params.Thinking.OfAdaptive != nil { + t.Error("should not set adaptive thinking for unknown level") + } +} + +func TestLevelToBudget(t *testing.T) { + tests := []struct { + name string + level string + want int + }{ + {"low", "low", 4096}, + {"medium", "medium", 16384}, + {"high", "high", 32000}, + {"xhigh", "xhigh", 64000}, + {"off", "off", 0}, + {"empty", "", 0}, + } + + for _, tt := range tests { + t.Run(tt.name, func(t *testing.T) { + if got := levelToBudget(tt.level); got != tt.want { + t.Errorf("levelToBudget(%q) = %d, want %d", tt.level, got, tt.want) + } + }) + } +} + +func TestBuildParams_ThinkingClearsTemperature(t *testing.T) { + msgs := []Message{{Role: "user", Content: "hello"}} + opts := map[string]any{ + "max_tokens": 200000, + "temperature": 0.8, + "thinking_level": "medium", + } + + params, err := buildParams(msgs, nil, "claude-sonnet-4-6", opts) + if err != nil { + t.Fatal(err) + } + + if params.Temperature.Valid() { + t.Error("temperature should be cleared when thinking_level is set") + } + if params.Thinking.OfEnabled == nil { + t.Fatal("expected enabled thinking") + } + if params.Thinking.OfEnabled.BudgetTokens != 16384 { + t.Errorf("budget_tokens = %d, want 16384", params.Thinking.OfEnabled.BudgetTokens) + } +} + +// unmarshalBlocks constructs []ContentBlockUnion via JSON round-trip so that +// the internal JSON.raw field is populated (required by AsText/AsThinking). +func unmarshalBlocks(t *testing.T, jsonStr string) []anthropic.ContentBlockUnion { + t.Helper() + var blocks []anthropic.ContentBlockUnion + if err := json.Unmarshal([]byte(jsonStr), &blocks); err != nil { + t.Fatalf("unmarshalBlocks: %v", err) + } + return blocks +} + +func TestParseResponse_ThinkingBlock(t *testing.T) { + resp := &anthropic.Message{ + Content: unmarshalBlocks(t, `[ + {"type":"thinking","thinking":"Let me reason step by step...","signature":"sig"}, + {"type":"text","text":"The answer is 42."} + ]`), + StopReason: anthropic.StopReasonEndTurn, + } + + result := parseResponse(resp) + + if result.Reasoning != "Let me reason step by step..." { + t.Errorf("Reasoning = %q, want thinking content", result.Reasoning) + } + if result.Content != "The answer is 42." { + t.Errorf("Content = %q, want text content", result.Content) + } + if result.FinishReason != "stop" { + t.Errorf("FinishReason = %q, want stop", result.FinishReason) + } +} + +func TestParseResponse_NoThinkingBlock(t *testing.T) { + resp := &anthropic.Message{ + Content: unmarshalBlocks(t, `[ + {"type":"text","text":"Just a normal response."} + ]`), + StopReason: anthropic.StopReasonEndTurn, + } + + result := parseResponse(resp) + + if result.Reasoning != "" { + t.Errorf("Reasoning = %q, want empty", result.Reasoning) + } + if result.Content != "Just a normal response." { + t.Errorf("Content = %q, want text content", result.Content) + } +} + +func TestBuildParams_NoThinkingKeepsTemperature(t *testing.T) { + msgs := []Message{{Role: "user", Content: "hello"}} + opts := map[string]any{ + "temperature": 0.8, + } + + params, err := buildParams(msgs, nil, "claude-sonnet-4-6", opts) + if err != nil { + t.Fatal(err) + } + + if !params.Temperature.Valid() { + t.Error("temperature should be preserved when thinking is not set") + } + if params.Temperature.Value != 0.8 { + t.Errorf("temperature = %f, want 0.8", params.Temperature.Value) + } +} diff --git a/pkg/providers/antigravity_provider.go b/pkg/providers/antigravity_provider.go index d4ee528b7..8a1890212 100644 --- a/pkg/providers/antigravity_provider.go +++ b/pkg/providers/antigravity_provider.go @@ -640,7 +640,10 @@ func FetchAntigravityProjectID(accessToken string) (string, error) { } defer resp.Body.Close() - body, _ := io.ReadAll(resp.Body) + body, err := io.ReadAll(resp.Body) + if err != nil { + return "", fmt.Errorf("reading loadCodeAssist response: %w", err) + } if resp.StatusCode != http.StatusOK { return "", fmt.Errorf("loadCodeAssist failed: %s", string(body)) } @@ -681,7 +684,10 @@ func FetchAntigravityModels(accessToken, projectID string) ([]AntigravityModelIn } defer resp.Body.Close() - body, _ := io.ReadAll(resp.Body) + body, err := io.ReadAll(resp.Body) + if err != nil { + return nil, fmt.Errorf("reading fetchAvailableModels response: %w", err) + } if resp.StatusCode != http.StatusOK { return nil, fmt.Errorf( "fetchAvailableModels failed (HTTP %d): %s", diff --git a/pkg/providers/factory.go b/pkg/providers/factory.go index 5b3e42b9e..a0d09a835 100644 --- a/pkg/providers/factory.go +++ b/pkg/providers/factory.go @@ -181,6 +181,15 @@ func resolveProviderSelection(cfg *config.Config) (providerSelection, error) { sel.model = "deepseek-chat" } } + case "avian": + if cfg.Providers.Avian.APIKey != "" { + sel.apiKey = cfg.Providers.Avian.APIKey + sel.apiBase = cfg.Providers.Avian.APIBase + sel.proxy = cfg.Providers.Avian.Proxy + if sel.apiBase == "" { + sel.apiBase = "https://api.avian.io/v1" + } + } case "mistral": if cfg.Providers.Mistral.APIKey != "" { sel.apiKey = cfg.Providers.Mistral.APIKey @@ -300,6 +309,13 @@ func resolveProviderSelection(cfg *config.Config) (providerSelection, error) { if sel.apiBase == "" { sel.apiBase = "https://api.mistral.ai/v1" } + case strings.HasPrefix(model, "avian/") && cfg.Providers.Avian.APIKey != "": + sel.apiKey = cfg.Providers.Avian.APIKey + sel.apiBase = cfg.Providers.Avian.APIBase + sel.proxy = cfg.Providers.Avian.Proxy + if sel.apiBase == "" { + sel.apiBase = "https://api.avian.io/v1" + } case cfg.Providers.VLLM.APIBase != "": sel.apiKey = cfg.Providers.VLLM.APIKey sel.apiBase = cfg.Providers.VLLM.APIBase diff --git a/pkg/providers/factory_provider.go b/pkg/providers/factory_provider.go index 155317a3b..c05fb0ad4 100644 --- a/pkg/providers/factory_provider.go +++ b/pkg/providers/factory_provider.go @@ -94,7 +94,7 @@ func CreateProviderFromConfig(cfg *config.ModelConfig) (LLMProvider, string, err case "litellm", "openrouter", "groq", "zhipu", "gemini", "nvidia", "ollama", "moonshot", "shengsuanyun", "deepseek", "cerebras", - "volcengine", "vllm", "qwen", "mistral": + "volcengine", "vllm", "qwen", "mistral", "avian": // All other OpenAI-compatible HTTP providers if cfg.APIKey == "" && cfg.APIBase == "" { return nil, "", fmt.Errorf("api_key or api_base is required for HTTP-based protocol %q", protocol) @@ -208,6 +208,8 @@ func getDefaultAPIBase(protocol string) string { return "http://localhost:8000/v1" case "mistral": return "https://api.mistral.ai/v1" + case "avian": + return "https://api.avian.io/v1" default: return "" } diff --git a/pkg/providers/openai_compat/provider.go b/pkg/providers/openai_compat/provider.go index ff9109e96..1904ee153 100644 --- a/pkg/providers/openai_compat/provider.go +++ b/pkg/providers/openai_compat/provider.go @@ -323,12 +323,14 @@ func serializeMessages(messages []Message) []any { }) } for _, mediaURL := range m.Media { - parts = append(parts, map[string]any{ - "type": "image_url", - "image_url": map[string]any{ - "url": mediaURL, - }, - }) + if strings.HasPrefix(mediaURL, "data:image/") { + parts = append(parts, map[string]any{ + "type": "image_url", + "image_url": map[string]any{ + "url": mediaURL, + }, + }) + } } msg := map[string]any{ diff --git a/pkg/providers/types.go b/pkg/providers/types.go index f0c168bc6..68bbd1e65 100644 --- a/pkg/providers/types.go +++ b/pkg/providers/types.go @@ -37,6 +37,13 @@ type StatefulProvider interface { Close() } +// ThinkingCapable is an optional interface for providers that support +// extended thinking (e.g. Anthropic). Used by the agent loop to warn +// when thinking_level is configured but the active provider cannot use it. +type ThinkingCapable interface { + SupportsThinking() bool +} + // FailoverReason classifies why an LLM request failed for fallback decisions. type FailoverReason string diff --git a/pkg/routing/classifier.go b/pkg/routing/classifier.go new file mode 100644 index 000000000..8cddaf069 --- /dev/null +++ b/pkg/routing/classifier.go @@ -0,0 +1,80 @@ +package routing + +// Classifier evaluates a feature set and returns a complexity score in [0, 1]. +// A higher score indicates a more complex task that benefits from a heavy model. +// The score is compared against the configured threshold: score >= threshold selects +// the primary (heavy) model; score < threshold selects the light model. +// +// Classifier is an interface so that future implementations (ML-based, embedding-based, +// or any other approach) can be swapped in without changing routing infrastructure. +type Classifier interface { + Score(f Features) float64 +} + +// RuleClassifier is the v1 implementation. +// It uses a weighted sum of structural signals with no external dependencies, +// no API calls, and sub-microsecond latency. The raw sum is capped at 1.0 so +// that the returned score always falls within the [0, 1] contract. +// +// Individual weights (multiple signals can fire simultaneously): +// +// token > 200 (≈600 chars): 0.35 — very long prompts are almost always complex +// token 50-200: 0.15 — medium length; may or may not be complex +// code block present: 0.40 — coding tasks need the heavy model +// tool calls > 3 (recent): 0.25 — dense tool usage signals an agentic workflow +// tool calls 1-3 (recent): 0.10 — some tool activity +// conversation depth > 10: 0.10 — long sessions carry implicit complexity +// attachments present: 1.00 — hard gate; multi-modal always needs heavy model +// +// Default threshold is 0.35, so: +// - Pure greetings / trivial Q&A: 0.00 → light ✓ +// - Medium prose message (50–200 tokens): 0.15 → light ✓ +// - Message with code block: 0.40 → heavy ✓ +// - Long message (>200 tokens): 0.35 → heavy ✓ +// - Active tool session + medium message: 0.25 → light (acceptable) +// - Any message with an image/audio attachment: 1.00 → heavy ✓ +type RuleClassifier struct{} + +// Score computes the complexity score for the given feature set. +// The returned value is in [0, 1]. Attachments short-circuit to 1.0. +func (c *RuleClassifier) Score(f Features) float64 { + // Hard gate: multi-modal inputs always require the heavy model. + if f.HasAttachments { + return 1.0 + } + + var score float64 + + // Token estimate — primary verbosity signal + switch { + case f.TokenEstimate > 200: + score += 0.35 + case f.TokenEstimate > 50: + score += 0.15 + } + + // Fenced code blocks — strongest indicator of a coding/technical task + if f.CodeBlockCount > 0 { + score += 0.40 + } + + // Recent tool call density — indicates an ongoing agentic workflow + switch { + case f.RecentToolCalls > 3: + score += 0.25 + case f.RecentToolCalls > 0: + score += 0.10 + } + + // Conversation depth — accumulated context implies compound task + if f.ConversationDepth > 10 { + score += 0.10 + } + + // Cap at 1.0 to honor the [0, 1] contract even when multiple signals fire + // simultaneously (e.g., long message + code block + tool chain = 1.10 raw). + if score > 1.0 { + score = 1.0 + } + return score +} diff --git a/pkg/routing/features.go b/pkg/routing/features.go new file mode 100644 index 000000000..c371e21aa --- /dev/null +++ b/pkg/routing/features.go @@ -0,0 +1,127 @@ +package routing + +import ( + "strings" + "unicode/utf8" + + "github.com/sipeed/picoclaw/pkg/providers" +) + +// lookbackWindow is the number of recent history entries scanned for tool calls. +// Six entries covers roughly one full tool-use round-trip (user → assistant+tool_call → tool_result → assistant). +const lookbackWindow = 6 + +// Features holds the structural signals extracted from a message and its session context. +// Every dimension is language-agnostic by construction — no keyword or pattern matching +// against natural-language content. This ensures consistent routing for all locales. +type Features struct { + // TokenEstimate is a proxy for token count. + // CJK runes count as 1 token each; non-CJK runes as 0.25 tokens each. + // This avoids API calls while giving accurate estimates for all scripts. + TokenEstimate int + + // CodeBlockCount is the number of fenced code blocks (``` pairs) in the message. + // Coding tasks almost always require the heavy model. + CodeBlockCount int + + // RecentToolCalls is the count of tool_call messages in the last lookbackWindow + // history entries. A high density indicates an active agentic workflow. + RecentToolCalls int + + // ConversationDepth is the total number of messages in the session history. + // Deep sessions tend to carry implicit complexity built up over many turns. + ConversationDepth int + + // HasAttachments is true when the message appears to contain media (images, + // audio, video). Multi-modal inputs require vision-capable heavy models. + HasAttachments bool +} + +// ExtractFeatures computes the structural feature vector for a message. +// It is a pure function with no side effects and zero allocations beyond +// the returned struct. +func ExtractFeatures(msg string, history []providers.Message) Features { + return Features{ + TokenEstimate: estimateTokens(msg), + CodeBlockCount: countCodeBlocks(msg), + RecentToolCalls: countRecentToolCalls(history), + ConversationDepth: len(history), + HasAttachments: hasAttachments(msg), + } +} + +// estimateTokens returns a token count proxy that handles both CJK and Latin text. +// CJK runes (U+2E80–U+9FFF, U+F900–U+FAFF, U+AC00–U+D7AF) map to roughly one +// token each, while non-CJK runes average ~0.25 tokens/rune (≈4 chars per token +// for English). Splitting the count this way avoids the 3x underestimation that a +// flat rune_count/3 would produce for Chinese, Japanese, and Korean text. +func estimateTokens(msg string) int { + total := utf8.RuneCountInString(msg) + if total == 0 { + return 0 + } + cjk := 0 + for _, r := range msg { + if r >= 0x2E80 && r <= 0x9FFF || r >= 0xF900 && r <= 0xFAFF || r >= 0xAC00 && r <= 0xD7AF { + cjk++ + } + } + return cjk + (total-cjk)/4 +} + +// countCodeBlocks counts the number of complete fenced code blocks. +// Each ``` delimiter increments a counter; pairs of delimiters form one block. +// An unclosed opening fence (odd count) is treated as zero complete blocks +// since it may just be an inline code span or a typo. +func countCodeBlocks(msg string) int { + n := strings.Count(msg, "```") + return n / 2 +} + +// countRecentToolCalls counts messages with tool calls in the last lookbackWindow +// entries of history. It examines the ToolCalls field rather than parsing +// the content string, so it is robust to any message format. +func countRecentToolCalls(history []providers.Message) int { + start := len(history) - lookbackWindow + if start < 0 { + start = 0 + } + + count := 0 + for _, msg := range history[start:] { + if len(msg.ToolCalls) > 0 { + count += len(msg.ToolCalls) + } + } + return count +} + +// hasAttachments returns true when the message content contains embedded media. +// It checks for base64 data URIs (data:image/, data:audio/, data:video/) and +// common image/audio URL extensions. This is intentionally conservative — +// false negatives (missing an attachment) just mean the routing falls back to +// the primary model anyway. +func hasAttachments(msg string) bool { + lower := strings.ToLower(msg) + + // Base64 data URIs embedded directly in the message + if strings.Contains(lower, "data:image/") || + strings.Contains(lower, "data:audio/") || + strings.Contains(lower, "data:video/") { + return true + } + + // Common image/audio extensions in URLs or file references + mediaExts := []string{ + ".jpg", ".jpeg", ".png", ".gif", ".webp", ".bmp", + ".mp3", ".wav", ".ogg", ".m4a", ".flac", + ".mp4", ".avi", ".mov", ".webm", + } + for _, ext := range mediaExts { + if strings.Contains(lower, ext) { + return true + } + } + + return false +} diff --git a/pkg/routing/router.go b/pkg/routing/router.go new file mode 100644 index 000000000..b1fa347e9 --- /dev/null +++ b/pkg/routing/router.go @@ -0,0 +1,82 @@ +package routing + +import ( + "github.com/sipeed/picoclaw/pkg/providers" +) + +// defaultThreshold is used when the config threshold is zero or negative. +// At 0.35 a message needs at least one strong signal (code block, long text, +// or an attachment) before the heavy model is chosen. +const defaultThreshold = 0.35 + +// RouterConfig holds the validated model routing settings. +// It mirrors config.RoutingConfig but lives in pkg/routing to keep the +// dependency graph simple: pkg/agent resolves config → routing, not the reverse. +type RouterConfig struct { + // LightModel is the model_name (from model_list) used for simple tasks. + LightModel string + + // Threshold is the complexity score cutoff in [0, 1]. + // score >= Threshold → primary (heavy) model. + // score < Threshold → light model. + Threshold float64 +} + +// Router selects the appropriate model tier for each incoming message. +// It is safe for concurrent use from multiple goroutines. +type Router struct { + cfg RouterConfig + classifier Classifier +} + +// New creates a Router with the given config and the default RuleClassifier. +// If cfg.Threshold is zero or negative, defaultThreshold (0.35) is used. +func New(cfg RouterConfig) *Router { + if cfg.Threshold <= 0 { + cfg.Threshold = defaultThreshold + } + return &Router{ + cfg: cfg, + classifier: &RuleClassifier{}, + } +} + +// newWithClassifier creates a Router with a custom Classifier. +// Intended for unit tests that need to inject a deterministic scorer. +func newWithClassifier(cfg RouterConfig, c Classifier) *Router { + if cfg.Threshold <= 0 { + cfg.Threshold = defaultThreshold + } + return &Router{cfg: cfg, classifier: c} +} + +// SelectModel returns the model to use for this conversation turn along with +// the computed complexity score (for logging and debugging). +// +// - If score < cfg.Threshold: returns (cfg.LightModel, true, score) +// - Otherwise: returns (primaryModel, false, score) +// +// The caller is responsible for resolving the returned model name into +// provider candidates (see AgentInstance.LightCandidates). +func (r *Router) SelectModel( + msg string, + history []providers.Message, + primaryModel string, +) (model string, usedLight bool, score float64) { + features := ExtractFeatures(msg, history) + score = r.classifier.Score(features) + if score < r.cfg.Threshold { + return r.cfg.LightModel, true, score + } + return primaryModel, false, score +} + +// LightModel returns the configured light model name. +func (r *Router) LightModel() string { + return r.cfg.LightModel +} + +// Threshold returns the complexity threshold in use. +func (r *Router) Threshold() float64 { + return r.cfg.Threshold +} diff --git a/pkg/routing/router_test.go b/pkg/routing/router_test.go new file mode 100644 index 000000000..2824d10ab --- /dev/null +++ b/pkg/routing/router_test.go @@ -0,0 +1,414 @@ +package routing + +import ( + "strings" + "testing" + + "github.com/sipeed/picoclaw/pkg/providers" +) + +// ── ExtractFeatures ────────────────────────────────────────────────────────── + +func TestExtractFeatures_EmptyMessage(t *testing.T) { + f := ExtractFeatures("", nil) + if f.TokenEstimate != 0 { + t.Errorf("TokenEstimate: got %d, want 0", f.TokenEstimate) + } + if f.CodeBlockCount != 0 { + t.Errorf("CodeBlockCount: got %d, want 0", f.CodeBlockCount) + } + if f.RecentToolCalls != 0 { + t.Errorf("RecentToolCalls: got %d, want 0", f.RecentToolCalls) + } + if f.ConversationDepth != 0 { + t.Errorf("ConversationDepth: got %d, want 0", f.ConversationDepth) + } + if f.HasAttachments { + t.Error("HasAttachments: got true, want false") + } +} + +func TestExtractFeatures_TokenEstimate(t *testing.T) { + // 30 ASCII runes: 0 CJK + 30/4 = 7 tokens + msg := strings.Repeat("a", 30) + f := ExtractFeatures(msg, nil) + if f.TokenEstimate != 7 { + t.Errorf("TokenEstimate: got %d, want 7", f.TokenEstimate) + } +} + +func TestExtractFeatures_TokenEstimate_CJK(t *testing.T) { + // 9 CJK runes → 9 tokens (each CJK rune ≈ 1 token). + // Using a rune slice literal avoids CJK string literals in source. + msg := string([]rune{ + 0x4F60, 0x597D, 0x4E16, 0x754C, + 0x4F60, 0x597D, 0x4E16, 0x754C, + 0x4F60, + }) + f := ExtractFeatures(msg, nil) + if f.TokenEstimate != 9 { + t.Errorf("CJK TokenEstimate: got %d, want 9", f.TokenEstimate) + } +} + +func TestExtractFeatures_TokenEstimate_Mixed(t *testing.T) { + // Mixed: 4 CJK runes + 8 ASCII runes → 4 + 8/4 = 6 tokens. + msg := string([]rune{0x4F60, 0x597D, 0x4E16, 0x754C}) + "hello ok" + f := ExtractFeatures(msg, nil) + if f.TokenEstimate != 6 { + t.Errorf("Mixed TokenEstimate: got %d, want 6", f.TokenEstimate) + } +} + +func TestExtractFeatures_CodeBlocks(t *testing.T) { + cases := []struct { + msg string + want int + }{ + {"no code here", 0}, + {"```go\nfmt.Println()\n```", 1}, + {"```python\npass\n```\n```js\nconsole.log()\n```", 2}, + {"```unclosed", 0}, // odd number of fences = 0 complete blocks + } + for _, tc := range cases { + f := ExtractFeatures(tc.msg, nil) + if f.CodeBlockCount != tc.want { + t.Errorf("msg=%q: CodeBlockCount got %d, want %d", tc.msg, f.CodeBlockCount, tc.want) + } + } +} + +func TestExtractFeatures_RecentToolCalls(t *testing.T) { + // History longer than lookbackWindow — only last lookbackWindow entries count. + history := make([]providers.Message, 10) + // Put 2 tool calls at positions 8 and 9 (within the last 6) + history[8] = providers.Message{Role: "assistant", ToolCalls: []providers.ToolCall{{Name: "exec"}}} + history[9] = providers.Message{ + Role: "assistant", + ToolCalls: []providers.ToolCall{{Name: "read_file"}, {Name: "write_file"}}, + } + // Position 3 is outside the lookback window and must NOT be counted + history[3] = providers.Message{Role: "assistant", ToolCalls: []providers.ToolCall{{Name: "old_tool"}}} + + f := ExtractFeatures("test", history) + // 1 (position 8) + 2 (position 9) = 3 + if f.RecentToolCalls != 3 { + t.Errorf("RecentToolCalls: got %d, want 3", f.RecentToolCalls) + } +} + +func TestExtractFeatures_ConversationDepth(t *testing.T) { + history := make([]providers.Message, 7) + f := ExtractFeatures("msg", history) + if f.ConversationDepth != 7 { + t.Errorf("ConversationDepth: got %d, want 7", f.ConversationDepth) + } +} + +func TestExtractFeatures_HasAttachments_DataURI(t *testing.T) { + cases := []struct { + msg string + want bool + }{ + {"plain text", false}, + {"here is an image: data:image/png;base64,abc123", true}, + {"audio: data:audio/mp3;base64,xyz", true}, + {"video: data:video/mp4;base64,xyz", true}, + } + for _, tc := range cases { + f := ExtractFeatures(tc.msg, nil) + if f.HasAttachments != tc.want { + t.Errorf("msg=%q: HasAttachments got %v, want %v", tc.msg, f.HasAttachments, tc.want) + } + } +} + +func TestExtractFeatures_HasAttachments_Extension(t *testing.T) { + cases := []struct { + msg string + want bool + }{ + {"check out photo.jpg", true}, + {"see screenshot.png", true}, + {"listen to audio.mp3", true}, + {"watch clip.mp4", true}, + {"just a .go file", false}, + {"document.pdf", false}, // pdf is not in the media list + } + for _, tc := range cases { + f := ExtractFeatures(tc.msg, nil) + if f.HasAttachments != tc.want { + t.Errorf("msg=%q: HasAttachments got %v, want %v", tc.msg, f.HasAttachments, tc.want) + } + } +} + +// ── RuleClassifier ─────────────────────────────────────────────────────────── + +func TestRuleClassifier_ZeroFeatures(t *testing.T) { + c := &RuleClassifier{} + score := c.Score(Features{}) + if score != 0.0 { + t.Errorf("zero features: got %f, want 0.0", score) + } +} + +func TestRuleClassifier_AttachmentsHardGate(t *testing.T) { + c := &RuleClassifier{} + score := c.Score(Features{HasAttachments: true}) + if score != 1.0 { + t.Errorf("attachments: got %f, want 1.0", score) + } +} + +func TestRuleClassifier_CodeBlockAlone(t *testing.T) { + c := &RuleClassifier{} + // Code block alone = 0.40, above default threshold 0.35 + score := c.Score(Features{CodeBlockCount: 1}) + if score < 0.35 { + t.Errorf("code block: score %f is below default threshold 0.35", score) + } +} + +func TestRuleClassifier_LongMessage(t *testing.T) { + c := &RuleClassifier{} + // >200 tokens = 0.35, exactly at default threshold → heavy + score := c.Score(Features{TokenEstimate: 250}) + if score < 0.35 { + t.Errorf("long message: score %f is below default threshold 0.35", score) + } +} + +func TestRuleClassifier_MediumMessage(t *testing.T) { + c := &RuleClassifier{} + // 50-200 tokens = 0.15, below threshold → light + score := c.Score(Features{TokenEstimate: 100}) + if score >= 0.35 { + t.Errorf("medium message: score %f should be below default threshold 0.35", score) + } +} + +func TestRuleClassifier_ShortMessage(t *testing.T) { + c := &RuleClassifier{} + // <50 tokens, no other signals = 0.0 → light + score := c.Score(Features{TokenEstimate: 10}) + if score != 0.0 { + t.Errorf("short message: got %f, want 0.0", score) + } +} + +func TestRuleClassifier_ToolCallDensity(t *testing.T) { + c := &RuleClassifier{} + + scoreNone := c.Score(Features{RecentToolCalls: 0}) + scoreLow := c.Score(Features{RecentToolCalls: 2}) + scoreHigh := c.Score(Features{RecentToolCalls: 5}) + + if scoreNone != 0.0 { + t.Errorf("no tools: got %f, want 0.0", scoreNone) + } + if scoreLow <= scoreNone { + t.Errorf("low tools should score higher than none: %f vs %f", scoreLow, scoreNone) + } + if scoreHigh <= scoreLow { + t.Errorf("high tools should score higher than low: %f vs %f", scoreHigh, scoreLow) + } +} + +func TestRuleClassifier_DeepConversation(t *testing.T) { + c := &RuleClassifier{} + shallow := c.Score(Features{ConversationDepth: 5}) + deep := c.Score(Features{ConversationDepth: 15}) + if deep <= shallow { + t.Errorf("deep conversation should score higher: %f vs %f", deep, shallow) + } +} + +func TestRuleClassifier_ScoreDoesNotExceedOne(t *testing.T) { + c := &RuleClassifier{} + // Max all signals simultaneously + f := Features{ + TokenEstimate: 500, + CodeBlockCount: 3, + RecentToolCalls: 10, + ConversationDepth: 20, + } + score := c.Score(f) + if score > 1.0 { + t.Errorf("score %f exceeds 1.0", score) + } +} + +// ── Router ─────────────────────────────────────────────────────────────────── + +func TestRouter_DefaultThreshold(t *testing.T) { + r := New(RouterConfig{LightModel: "gemini-flash"}) + if r.Threshold() != defaultThreshold { + t.Errorf("default threshold: got %f, want %f", r.Threshold(), defaultThreshold) + } +} + +func TestRouter_NegativeThresholdFallsBackToDefault(t *testing.T) { + r := New(RouterConfig{LightModel: "gemini-flash", Threshold: -0.1}) + if r.Threshold() != defaultThreshold { + t.Errorf("negative threshold: got %f, want %f", r.Threshold(), defaultThreshold) + } +} + +func TestRouter_SelectModel_SimpleMessageUsesLight(t *testing.T) { + r := New(RouterConfig{LightModel: "gemini-flash", Threshold: 0.35}) + msg := "hi" + model, usedLight, _ := r.SelectModel(msg, nil, "claude-sonnet-4-6") + if !usedLight { + t.Error("simple message: expected light model to be selected") + } + if model != "gemini-flash" { + t.Errorf("simple message: model got %q, want %q", model, "gemini-flash") + } +} + +func TestRouter_SelectModel_CodeBlockUsesPrimary(t *testing.T) { + r := New(RouterConfig{LightModel: "gemini-flash", Threshold: 0.35}) + msg := "```go\nfmt.Println(\"hello\")\n```" + model, usedLight, _ := r.SelectModel(msg, nil, "claude-sonnet-4-6") + if usedLight { + t.Error("code block: expected primary model to be selected") + } + if model != "claude-sonnet-4-6" { + t.Errorf("code block: model got %q, want %q", model, "claude-sonnet-4-6") + } +} + +func TestRouter_SelectModel_AttachmentUsesPrimary(t *testing.T) { + r := New(RouterConfig{LightModel: "gemini-flash", Threshold: 0.35}) + msg := "can you analyze this? data:image/png;base64,abc123" + model, usedLight, _ := r.SelectModel(msg, nil, "claude-sonnet-4-6") + if usedLight { + t.Error("attachment: expected primary model to be selected") + } + if model != "claude-sonnet-4-6" { + t.Errorf("attachment: model got %q, want %q", model, "claude-sonnet-4-6") + } +} + +func TestRouter_SelectModel_LongMessageUsesPrimary(t *testing.T) { + r := New(RouterConfig{LightModel: "gemini-flash", Threshold: 0.35}) + // >200 token estimate: 210 * 3 = 630 chars + msg := strings.Repeat("word ", 210) + model, usedLight, _ := r.SelectModel(msg, nil, "claude-sonnet-4-6") + if usedLight { + t.Error("long message: expected primary model to be selected") + } + if model != "claude-sonnet-4-6" { + t.Errorf("long message: model got %q, want %q", model, "claude-sonnet-4-6") + } +} + +func TestRouter_SelectModel_DeepToolChainUsesLight(t *testing.T) { + // Tool calls alone (0.25) don't cross the 0.35 threshold — acceptable behavior. + // Routing is conservative: only promote to heavy when the signal is unambiguous. + r := New(RouterConfig{LightModel: "gemini-flash", Threshold: 0.35}) + history := []providers.Message{ + {Role: "assistant", ToolCalls: []providers.ToolCall{{Name: "read_file"}, {Name: "write_file"}}}, + {Role: "assistant", ToolCalls: []providers.ToolCall{{Name: "exec"}, {Name: "search"}}}, + } + msg := "ok" + _, usedLight, _ := r.SelectModel(msg, history, "claude-sonnet-4-6") + if !usedLight { + t.Error("short message + moderate tool calls: expected light model (score 0.20 < 0.35)") + } +} + +func TestRouter_SelectModel_ToolChainPlusMediumUsesHeavy(t *testing.T) { + // Tool calls (0.25) + medium message (0.15) = 0.40 >= 0.35 → heavy + r := New(RouterConfig{LightModel: "gemini-flash", Threshold: 0.35}) + history := []providers.Message{ + {Role: "assistant", ToolCalls: []providers.ToolCall{ + {Name: "a"}, {Name: "b"}, {Name: "c"}, {Name: "d"}, + }}, + } + // ~55 tokens * 3 = 165 chars + msg := strings.Repeat("word ", 55) + _, usedLight, _ := r.SelectModel(msg, history, "claude-sonnet-4-6") + if usedLight { + t.Error("tool chain + medium message: expected primary model (score >= 0.35)") + } +} + +func TestRouter_SelectModel_CustomThreshold(t *testing.T) { + // Very low threshold: even a short message triggers heavy model + r := New(RouterConfig{LightModel: "gemini-flash", Threshold: 0.05}) + msg := strings.Repeat("word ", 55) // medium message → 0.15 >= 0.05 + _, usedLight, _ := r.SelectModel(msg, nil, "claude-sonnet-4-6") + if usedLight { + t.Error("low threshold: medium message should use primary model") + } +} + +func TestRouter_SelectModel_HighThreshold(t *testing.T) { + // Very high threshold: even code blocks route to light + r := New(RouterConfig{LightModel: "gemini-flash", Threshold: 0.99}) + msg := "```go\nfmt.Println()\n```" + _, usedLight, _ := r.SelectModel(msg, nil, "claude-sonnet-4-6") + if !usedLight { + t.Error("very high threshold: code block (0.40) should route to light model") + } +} + +func TestRouter_LightModel(t *testing.T) { + r := New(RouterConfig{LightModel: "my-fast-model", Threshold: 0.35}) + if r.LightModel() != "my-fast-model" { + t.Errorf("LightModel: got %q, want %q", r.LightModel(), "my-fast-model") + } +} + +// ── newWithClassifier (internal testing hook) ───────────────────────────────── + +type fixedScoreClassifier struct{ score float64 } + +func (f *fixedScoreClassifier) Score(_ Features) float64 { return f.score } + +func TestRouter_CustomClassifier_LowScore_SelectsLight(t *testing.T) { + r := newWithClassifier( + RouterConfig{LightModel: "light", Threshold: 0.5}, + &fixedScoreClassifier{score: 0.2}, + ) + _, usedLight, _ := r.SelectModel("anything", nil, "heavy") + if !usedLight { + t.Error("low score with custom classifier: expected light model") + } +} + +func TestRouter_CustomClassifier_HighScore_SelectsPrimary(t *testing.T) { + r := newWithClassifier( + RouterConfig{LightModel: "light", Threshold: 0.5}, + &fixedScoreClassifier{score: 0.8}, + ) + _, usedLight, _ := r.SelectModel("anything", nil, "heavy") + if usedLight { + t.Error("high score with custom classifier: expected primary model") + } +} + +func TestRouter_CustomClassifier_ExactThreshold_SelectsPrimary(t *testing.T) { + // score == threshold → primary (uses >= comparison) + r := newWithClassifier( + RouterConfig{LightModel: "light", Threshold: 0.5}, + &fixedScoreClassifier{score: 0.5}, + ) + _, usedLight, _ := r.SelectModel("anything", nil, "heavy") + if usedLight { + t.Error("score == threshold: expected primary model (>= threshold → primary)") + } +} + +func TestRouter_SelectModel_ReturnsScore(t *testing.T) { + r := newWithClassifier( + RouterConfig{LightModel: "light", Threshold: 0.5}, + &fixedScoreClassifier{score: 0.42}, + ) + _, _, score := r.SelectModel("anything", nil, "heavy") + if score != 0.42 { + t.Errorf("score: got %f, want 0.42", score) + } +} diff --git a/pkg/skills/clawhub_registry.go b/pkg/skills/clawhub_registry.go index f78197bbe..bd4bed8fb 100644 --- a/pkg/skills/clawhub_registry.go +++ b/pkg/skills/clawhub_registry.go @@ -259,15 +259,7 @@ func (c *ClawHubRegistry) DownloadAndInstall( } u.RawQuery = q.Encode() - req, err := http.NewRequestWithContext(ctx, "GET", u.String(), nil) - if err != nil { - return nil, fmt.Errorf("failed to create request: %w", err) - } - if c.authToken != "" { - req.Header.Set("Authorization", "Bearer "+c.authToken) - } - - tmpPath, err := utils.DownloadToFile(ctx, c.client, req, int64(c.maxZipSize)) + tmpPath, err := c.downloadToTempFileWithRetry(ctx, u.String()) if err != nil { return nil, fmt.Errorf("download failed: %w", err) } @@ -284,17 +276,12 @@ func (c *ClawHubRegistry) DownloadAndInstall( // --- HTTP helper --- func (c *ClawHubRegistry) doGet(ctx context.Context, urlStr string) ([]byte, error) { - req, err := http.NewRequestWithContext(ctx, "GET", urlStr, nil) + req, err := c.newGetRequest(ctx, urlStr, "application/json") if err != nil { return nil, err } - req.Header.Set("Accept", "application/json") - if c.authToken != "" { - req.Header.Set("Authorization", "Bearer "+c.authToken) - } - - resp, err := c.client.Do(req) + resp, err := utils.DoRequestWithRetry(c.client, req) if err != nil { return nil, err } @@ -312,3 +299,64 @@ func (c *ClawHubRegistry) doGet(ctx context.Context, urlStr string) ([]byte, err return body, nil } + +func (c *ClawHubRegistry) newGetRequest(ctx context.Context, urlStr, accept string) (*http.Request, error) { + req, err := http.NewRequestWithContext(ctx, http.MethodGet, urlStr, nil) + if err != nil { + return nil, err + } + req.Header.Set("Accept", accept) + if c.authToken != "" { + req.Header.Set("Authorization", "Bearer "+c.authToken) + } + return req, nil +} + +func (c *ClawHubRegistry) downloadToTempFileWithRetry(ctx context.Context, urlStr string) (string, error) { + req, err := c.newGetRequest(ctx, urlStr, "application/zip") + if err != nil { + return "", err + } + + resp, err := utils.DoRequestWithRetry(c.client, req) + if err != nil { + return "", err + } + defer resp.Body.Close() + + if resp.StatusCode < 200 || resp.StatusCode >= 300 { + errBody := make([]byte, 512) + n, _ := io.ReadFull(resp.Body, errBody) + return "", fmt.Errorf("HTTP %d: %s", resp.StatusCode, string(errBody[:n])) + } + + tmpFile, err := os.CreateTemp("", "picoclaw-dl-*") + if err != nil { + return "", fmt.Errorf("failed to create temp file: %w", err) + } + tmpPath := tmpFile.Name() + + cleanup := func() { + _ = tmpFile.Close() + _ = os.Remove(tmpPath) + } + + src := io.LimitReader(resp.Body, int64(c.maxZipSize)+1) + written, err := io.Copy(tmpFile, src) + if err != nil { + cleanup() + return "", fmt.Errorf("download write failed: %w", err) + } + + if written > int64(c.maxZipSize) { + cleanup() + return "", fmt.Errorf("download too large: %d bytes (max %d)", written, c.maxZipSize) + } + + if err := tmpFile.Close(); err != nil { + _ = os.Remove(tmpPath) + return "", fmt.Errorf("failed to close temp file: %w", err) + } + + return tmpPath, nil +} diff --git a/pkg/skills/clawhub_registry_test.go b/pkg/skills/clawhub_registry_test.go index 65ee638da..055da22dc 100644 --- a/pkg/skills/clawhub_registry_test.go +++ b/pkg/skills/clawhub_registry_test.go @@ -54,6 +54,39 @@ func TestClawHubRegistrySearch(t *testing.T) { assert.Equal(t, "clawhub", results[0].RegistryName) } +func TestClawHubRegistrySearchRetries429(t *testing.T) { + attempts := 0 + srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + attempts++ + if attempts == 1 { + w.Header().Set("Retry-After", "0") + w.WriteHeader(http.StatusTooManyRequests) + w.Write([]byte("rate limited")) + return + } + + slug := "github" + name := "GitHub Integration" + summary := "Interact with GitHub repos" + version := "1.0.0" + + json.NewEncoder(w).Encode(clawhubSearchResponse{ + Results: []clawhubSearchResult{ + {Score: 0.95, Slug: &slug, DisplayName: &name, Summary: &summary, Version: &version}, + }, + }) + })) + defer srv.Close() + + reg := newTestRegistry(srv.URL, "") + results, err := reg.Search(context.Background(), "github", 5) + + require.NoError(t, err) + require.Len(t, results, 1) + assert.Equal(t, 2, attempts) + assert.Equal(t, "github", results[0].Slug) +} + func TestClawHubRegistryGetSkillMeta(t *testing.T) { srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { assert.Equal(t, "/api/v1/skills/github", r.URL.Path) @@ -137,6 +170,54 @@ func TestClawHubRegistryDownloadAndInstall(t *testing.T) { assert.Contains(t, string(readmeContent), "# Test Skill") } +func TestClawHubRegistryDownloadAndInstallRetries429(t *testing.T) { + zipBuf := createTestZip(t, map[string]string{ + "SKILL.md": "---\nname: retry-skill\ndescription: A test\n---\nHello skill", + }) + + downloadAttempts := 0 + srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + switch r.URL.Path { + case "/api/v1/skills/retry-skill": + json.NewEncoder(w).Encode(clawhubSkillResponse{ + Slug: "retry-skill", + DisplayName: "Retry Skill", + Summary: "A retry test skill", + LatestVersion: &clawhubVersionInfo{Version: "1.0.0"}, + }) + case "/api/v1/download": + downloadAttempts++ + if downloadAttempts == 1 { + w.Header().Set("Retry-After", "0") + w.WriteHeader(http.StatusTooManyRequests) + w.Write([]byte("rate limited")) + return + } + assert.Equal(t, "retry-skill", r.URL.Query().Get("slug")) + w.Header().Set("Content-Type", "application/zip") + w.Write(zipBuf) + default: + w.WriteHeader(http.StatusNotFound) + } + })) + defer srv.Close() + + tmpDir := t.TempDir() + targetDir := filepath.Join(tmpDir, "retry-skill") + + reg := newTestRegistry(srv.URL, "") + result, err := reg.DownloadAndInstall(context.Background(), "retry-skill", "", targetDir) + + require.NoError(t, err) + require.NotNil(t, result) + assert.Equal(t, "1.0.0", result.Version) + assert.Equal(t, 2, downloadAttempts) + + skillContent, err := os.ReadFile(filepath.Join(targetDir, "SKILL.md")) + require.NoError(t, err) + assert.Contains(t, string(skillContent), "Hello skill") +} + func TestClawHubRegistryAuthToken(t *testing.T) { srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { authHeader := r.Header.Get("Authorization") diff --git a/pkg/tools/base.go b/pkg/tools/base.go index 770d8cb04..ec743e164 100644 --- a/pkg/tools/base.go +++ b/pkg/tools/base.go @@ -10,11 +10,38 @@ type Tool interface { Execute(ctx context.Context, args map[string]any) *ToolResult } -// ContextualTool is an optional interface that tools can implement -// to receive the current message context (channel, chatID) -type ContextualTool interface { - Tool - SetContext(channel, chatID string) +// --- Request-scoped tool context (channel / chatID) --- +// +// Carried via context.Value so that concurrent tool calls each receive +// their own immutable copy — no mutable state on singleton tool instances. +// +// Keys are unexported pointer-typed vars — guaranteed collision-free, +// and only accessible through the helper functions below. + +type toolCtxKey struct{ name string } + +var ( + ctxKeyChannel = &toolCtxKey{"channel"} + ctxKeyChatID = &toolCtxKey{"chatID"} +) + +// WithToolContext returns a child context carrying channel and chatID. +func WithToolContext(ctx context.Context, channel, chatID string) context.Context { + ctx = context.WithValue(ctx, ctxKeyChannel, channel) + ctx = context.WithValue(ctx, ctxKeyChatID, chatID) + return ctx +} + +// ToolChannel extracts the channel from ctx, or "" if unset. +func ToolChannel(ctx context.Context) string { + v, _ := ctx.Value(ctxKeyChannel).(string) + return v +} + +// ToolChatID extracts the chatID from ctx, or "" if unset. +func ToolChatID(ctx context.Context) string { + v, _ := ctx.Value(ctxKeyChatID).(string) + return v } // AsyncCallback is a function type that async tools use to notify completion. @@ -22,51 +49,36 @@ type ContextualTool interface { // // The ctx parameter allows the callback to be canceled if the agent is shutting down. // The result parameter contains the tool's execution result. -// -// Example usage in an async tool: -// -// func (t *MyAsyncTool) Execute(ctx context.Context, args map[string]interface{}) *ToolResult { -// // Start async work in background -// go func() { -// result := doAsyncWork() -// if t.callback != nil { -// t.callback(ctx, result) -// } -// }() -// return AsyncResult("Async task started") -// } type AsyncCallback func(ctx context.Context, result *ToolResult) -// AsyncTool is an optional interface that tools can implement to support +// AsyncExecutor is an optional interface that tools can implement to support // asynchronous execution with completion callbacks. // -// Async tools return immediately with an AsyncResult, then notify completion -// via the callback set by SetCallback. +// Unlike the old AsyncTool pattern (SetCallback + Execute), AsyncExecutor +// receives the callback as a parameter of ExecuteAsync. This eliminates the +// data race where concurrent calls could overwrite each other's callbacks +// on a shared tool instance. // // This is useful for: -// - Long-running operations that shouldn't block the agent loop -// - Subagent spawns that complete independently -// - Background tasks that need to report results later +// - Long-running operations that shouldn't block the agent loop +// - Subagent spawns that complete independently +// - Background tasks that need to report results later // // Example: // -// type SpawnTool struct { -// callback AsyncCallback -// } -// -// func (t *SpawnTool) SetCallback(cb AsyncCallback) { -// t.callback = cb -// } -// -// func (t *SpawnTool) Execute(ctx context.Context, args map[string]interface{}) *ToolResult { -// go t.runSubagent(ctx, args) +// func (t *SpawnTool) ExecuteAsync(ctx context.Context, args map[string]any, cb AsyncCallback) *ToolResult { +// go func() { +// result := t.runSubagent(ctx, args) +// if cb != nil { cb(ctx, result) } +// }() // return AsyncResult("Subagent spawned, will report back") // } -type AsyncTool interface { +type AsyncExecutor interface { Tool - // SetCallback registers a callback function to be invoked when the async operation completes. - // The callback will be called from a goroutine and should handle thread-safety if needed. - SetCallback(cb AsyncCallback) + // ExecuteAsync runs the tool asynchronously. The callback cb will be + // invoked (possibly from another goroutine) when the async operation + // completes. cb is guaranteed to be non-nil by the caller (registry). + ExecuteAsync(ctx context.Context, args map[string]any, cb AsyncCallback) *ToolResult } func ToolToSchema(tool Tool) map[string]any { diff --git a/pkg/tools/cron.go b/pkg/tools/cron.go index 6888d1326..6af0aa9e1 100644 --- a/pkg/tools/cron.go +++ b/pkg/tools/cron.go @@ -4,7 +4,6 @@ import ( "context" "fmt" "strings" - "sync" "time" "github.com/sipeed/picoclaw/pkg/bus" @@ -24,9 +23,6 @@ type CronTool struct { executor JobExecutor msgBus *bus.MessageBus execTool *ExecTool - channel string - chatID string - mu sync.RWMutex } // NewCronTool creates a new CronTool @@ -102,14 +98,6 @@ func (t *CronTool) Parameters() map[string]any { } } -// SetContext sets the current session context for job creation -func (t *CronTool) SetContext(channel, chatID string) { - t.mu.Lock() - defer t.mu.Unlock() - t.channel = channel - t.chatID = chatID -} - // Execute runs the tool with the given arguments func (t *CronTool) Execute(ctx context.Context, args map[string]any) *ToolResult { action, ok := args["action"].(string) @@ -119,7 +107,7 @@ func (t *CronTool) Execute(ctx context.Context, args map[string]any) *ToolResult switch action { case "add": - return t.addJob(args) + return t.addJob(ctx, args) case "list": return t.listJobs() case "remove": @@ -133,11 +121,9 @@ func (t *CronTool) Execute(ctx context.Context, args map[string]any) *ToolResult } } -func (t *CronTool) addJob(args map[string]any) *ToolResult { - t.mu.RLock() - channel := t.channel - chatID := t.chatID - t.mu.RUnlock() +func (t *CronTool) addJob(ctx context.Context, args map[string]any) *ToolResult { + channel := ToolChannel(ctx) + chatID := ToolChatID(ctx) if channel == "" || chatID == "" { return ErrorResult("no session context (channel/chat_id not set). Use this tool in an active conversation.") @@ -155,6 +141,12 @@ func (t *CronTool) addJob(args map[string]any) *ToolResult { everySeconds, hasEvery := args["every_seconds"].(float64) cronExpr, hasCron := args["cron_expr"].(string) + // Fix: type assertions return true for zero values, need additional validity checks + // This prevents LLMs that fill unused optional parameters with defaults (0) from triggering wrong type + hasAt = hasAt && atSeconds > 0 + hasEvery = hasEvery && everySeconds > 0 + hasCron = hasCron && cronExpr != "" + // Priority: at_seconds > every_seconds > cron_expr if hasAt { atMS := time.Now().UnixMilli() + int64(atSeconds)*1000 diff --git a/pkg/tools/message.go b/pkg/tools/message.go index 15ef4ff73..438ceeddd 100644 --- a/pkg/tools/message.go +++ b/pkg/tools/message.go @@ -3,15 +3,14 @@ package tools import ( "context" "fmt" + "sync/atomic" ) type SendCallback func(channel, chatID, content string) error type MessageTool struct { - sendCallback SendCallback - defaultChannel string - defaultChatID string - sentInRound bool // Tracks whether a message was sent in the current processing round + sendCallback SendCallback + sentInRound atomic.Bool // Tracks whether a message was sent in the current processing round } func NewMessageTool() *MessageTool { @@ -47,15 +46,15 @@ func (t *MessageTool) Parameters() map[string]any { } } -func (t *MessageTool) SetContext(channel, chatID string) { - t.defaultChannel = channel - t.defaultChatID = chatID - t.sentInRound = false // Reset send tracking for new processing round +// ResetSentInRound resets the per-round send tracker. +// Called by the agent loop at the start of each inbound message processing round. +func (t *MessageTool) ResetSentInRound() { + t.sentInRound.Store(false) } // HasSentInRound returns true if the message tool sent a message during the current round. func (t *MessageTool) HasSentInRound() bool { - return t.sentInRound + return t.sentInRound.Load() } func (t *MessageTool) SetSendCallback(callback SendCallback) { @@ -72,10 +71,10 @@ func (t *MessageTool) Execute(ctx context.Context, args map[string]any) *ToolRes chatID, _ := args["chat_id"].(string) if channel == "" { - channel = t.defaultChannel + channel = ToolChannel(ctx) } if chatID == "" { - chatID = t.defaultChatID + chatID = ToolChatID(ctx) } if channel == "" || chatID == "" { @@ -94,7 +93,7 @@ func (t *MessageTool) Execute(ctx context.Context, args map[string]any) *ToolRes } } - t.sentInRound = true + t.sentInRound.Store(true) // Silent: user already received the message directly return &ToolResult{ ForLLM: fmt.Sprintf("Message sent to %s:%s", channel, chatID), diff --git a/pkg/tools/message_test.go b/pkg/tools/message_test.go index 717c1117b..05630972e 100644 --- a/pkg/tools/message_test.go +++ b/pkg/tools/message_test.go @@ -8,7 +8,6 @@ import ( func TestMessageTool_Execute_Success(t *testing.T) { tool := NewMessageTool() - tool.SetContext("test-channel", "test-chat-id") var sentChannel, sentChatID, sentContent string tool.SetSendCallback(func(channel, chatID, content string) error { @@ -18,7 +17,7 @@ func TestMessageTool_Execute_Success(t *testing.T) { return nil }) - ctx := context.Background() + ctx := WithToolContext(context.Background(), "test-channel", "test-chat-id") args := map[string]any{ "content": "Hello, world!", } @@ -60,7 +59,6 @@ func TestMessageTool_Execute_Success(t *testing.T) { func TestMessageTool_Execute_WithCustomChannel(t *testing.T) { tool := NewMessageTool() - tool.SetContext("default-channel", "default-chat-id") var sentChannel, sentChatID string tool.SetSendCallback(func(channel, chatID, content string) error { @@ -69,7 +67,7 @@ func TestMessageTool_Execute_WithCustomChannel(t *testing.T) { return nil }) - ctx := context.Background() + ctx := WithToolContext(context.Background(), "default-channel", "default-chat-id") args := map[string]any{ "content": "Test message", "channel": "custom-channel", @@ -96,14 +94,13 @@ func TestMessageTool_Execute_WithCustomChannel(t *testing.T) { func TestMessageTool_Execute_SendFailure(t *testing.T) { tool := NewMessageTool() - tool.SetContext("test-channel", "test-chat-id") sendErr := errors.New("network error") tool.SetSendCallback(func(channel, chatID, content string) error { return sendErr }) - ctx := context.Background() + ctx := WithToolContext(context.Background(), "test-channel", "test-chat-id") args := map[string]any{ "content": "Test message", } @@ -133,9 +130,8 @@ func TestMessageTool_Execute_SendFailure(t *testing.T) { func TestMessageTool_Execute_MissingContent(t *testing.T) { tool := NewMessageTool() - tool.SetContext("test-channel", "test-chat-id") - ctx := context.Background() + ctx := WithToolContext(context.Background(), "test-channel", "test-chat-id") args := map[string]any{} // content missing result := tool.Execute(ctx, args) @@ -151,7 +147,7 @@ func TestMessageTool_Execute_MissingContent(t *testing.T) { func TestMessageTool_Execute_NoTargetChannel(t *testing.T) { tool := NewMessageTool() - // No SetContext called, so defaultChannel and defaultChatID are empty + // No WithToolContext — channel/chatID are empty tool.SetSendCallback(func(channel, chatID, content string) error { return nil @@ -175,10 +171,9 @@ func TestMessageTool_Execute_NoTargetChannel(t *testing.T) { func TestMessageTool_Execute_NotConfigured(t *testing.T) { tool := NewMessageTool() - tool.SetContext("test-channel", "test-chat-id") // No SetSendCallback called - ctx := context.Background() + ctx := WithToolContext(context.Background(), "test-channel", "test-chat-id") args := map[string]any{ "content": "Test message", } diff --git a/pkg/tools/registry.go b/pkg/tools/registry.go index 0ba983e02..ca8436c67 100644 --- a/pkg/tools/registry.go +++ b/pkg/tools/registry.go @@ -45,8 +45,9 @@ func (r *ToolRegistry) Execute(ctx context.Context, name string, args map[string } // ExecuteWithContext executes a tool with channel/chatID context and optional async callback. -// If the tool implements AsyncTool and a non-nil callback is provided, -// the callback will be set on the tool before execution. +// If the tool implements AsyncExecutor and a non-nil callback is provided, +// ExecuteAsync is called instead of Execute — the callback is a parameter, +// never stored as mutable state on the tool. func (r *ToolRegistry) ExecuteWithContext( ctx context.Context, name string, @@ -69,22 +70,23 @@ func (r *ToolRegistry) ExecuteWithContext( return ErrorResult(fmt.Sprintf("tool %q not found", name)).WithError(fmt.Errorf("tool not found")) } - // If tool implements ContextualTool, set context - if contextualTool, ok := tool.(ContextualTool); ok && channel != "" && chatID != "" { - contextualTool.SetContext(channel, chatID) - } + // Inject channel/chatID into ctx so tools read them via ToolChannel(ctx)/ToolChatID(ctx). + // Always inject — tools validate what they require. + ctx = WithToolContext(ctx, channel, chatID) - // If tool implements AsyncTool and callback is provided, set callback - if asyncTool, ok := tool.(AsyncTool); ok && asyncCallback != nil { - asyncTool.SetCallback(asyncCallback) - logger.DebugCF("tool", "Async callback injected", + // If tool implements AsyncExecutor and callback is provided, use ExecuteAsync. + // The callback is a call parameter, not mutable state on the tool instance. + var result *ToolResult + start := time.Now() + if asyncExec, ok := tool.(AsyncExecutor); ok && asyncCallback != nil { + logger.DebugCF("tool", "Executing async tool via ExecuteAsync", map[string]any{ "tool": name, }) + result = asyncExec.ExecuteAsync(ctx, args, asyncCallback) + } else { + result = tool.Execute(ctx, args) } - - start := time.Now() - result := tool.Execute(ctx, args) duration := time.Since(start) // Log based on result type diff --git a/pkg/tools/registry_test.go b/pkg/tools/registry_test.go index 8fe88ca78..92d7d5abd 100644 --- a/pkg/tools/registry_test.go +++ b/pkg/tools/registry_test.go @@ -25,24 +25,24 @@ func (m *mockRegistryTool) Execute(_ context.Context, _ map[string]any) *ToolRes return m.result } -type mockCtxTool struct { +type mockContextAwareTool struct { mockRegistryTool - channel string - chatID string + lastCtx context.Context } -func (m *mockCtxTool) SetContext(channel, chatID string) { - m.channel = channel - m.chatID = chatID +func (m *mockContextAwareTool) Execute(ctx context.Context, _ map[string]any) *ToolResult { + m.lastCtx = ctx + return m.result } type mockAsyncRegistryTool struct { mockRegistryTool - cb AsyncCallback + lastCB AsyncCallback } -func (m *mockAsyncRegistryTool) SetCallback(cb AsyncCallback) { - m.cb = cb +func (m *mockAsyncRegistryTool) ExecuteAsync(_ context.Context, args map[string]any, cb AsyncCallback) *ToolResult { + m.lastCB = cb + return m.result } // --- helpers --- @@ -136,34 +136,44 @@ func TestToolRegistry_Execute_NotFound(t *testing.T) { } } -func TestToolRegistry_ExecuteWithContext_ContextualTool(t *testing.T) { +func TestToolRegistry_ExecuteWithContext_InjectsToolContext(t *testing.T) { r := NewToolRegistry() - ct := &mockCtxTool{ + ct := &mockContextAwareTool{ mockRegistryTool: *newMockTool("ctx_tool", "needs context"), } r.Register(ct) r.ExecuteWithContext(context.Background(), "ctx_tool", nil, "telegram", "chat-42", nil) - if ct.channel != "telegram" { - t.Errorf("expected channel 'telegram', got %q", ct.channel) + if ct.lastCtx == nil { + t.Fatal("expected Execute to be called") } - if ct.chatID != "chat-42" { - t.Errorf("expected chatID 'chat-42', got %q", ct.chatID) + if got := ToolChannel(ct.lastCtx); got != "telegram" { + t.Errorf("expected channel 'telegram', got %q", got) + } + if got := ToolChatID(ct.lastCtx); got != "chat-42" { + t.Errorf("expected chatID 'chat-42', got %q", got) } } -func TestToolRegistry_ExecuteWithContext_SkipsEmptyContext(t *testing.T) { +func TestToolRegistry_ExecuteWithContext_EmptyContext(t *testing.T) { r := NewToolRegistry() - ct := &mockCtxTool{ + ct := &mockContextAwareTool{ mockRegistryTool: *newMockTool("ctx_tool", "needs context"), } r.Register(ct) r.ExecuteWithContext(context.Background(), "ctx_tool", nil, "", "", nil) - if ct.channel != "" || ct.chatID != "" { - t.Error("SetContext should not be called with empty channel/chatID") + if ct.lastCtx == nil { + t.Fatal("expected Execute to be called") + } + // Empty values are still injected; tools decide what to do with them. + if got := ToolChannel(ct.lastCtx); got != "" { + t.Errorf("expected empty channel, got %q", got) + } + if got := ToolChatID(ct.lastCtx); got != "" { + t.Errorf("expected empty chatID, got %q", got) } } @@ -179,14 +189,14 @@ func TestToolRegistry_ExecuteWithContext_AsyncCallback(t *testing.T) { cb := func(_ context.Context, _ *ToolResult) { called = true } result := r.ExecuteWithContext(context.Background(), "async_tool", nil, "", "", cb) - if at.cb == nil { - t.Error("expected SetCallback to have been called") + if at.lastCB == nil { + t.Error("expected ExecuteAsync to have received a callback") } if !result.Async { t.Error("expected async result") } - at.cb(context.Background(), SilentResult("done")) + at.lastCB(context.Background(), SilentResult("done")) if !called { t.Error("expected callback to be invoked") } diff --git a/pkg/tools/send_file.go b/pkg/tools/send_file.go new file mode 100644 index 000000000..1a03e58ed --- /dev/null +++ b/pkg/tools/send_file.go @@ -0,0 +1,150 @@ +package tools + +import ( + "context" + "fmt" + "mime" + "os" + "path/filepath" + "strings" + + "github.com/h2non/filetype" + + "github.com/sipeed/picoclaw/pkg/config" + "github.com/sipeed/picoclaw/pkg/media" +) + +// SendFileTool allows the LLM to send a local file (image, document, etc.) +// to the user on the current chat channel via the MediaStore pipeline. +type SendFileTool struct { + workspace string + restrict bool + maxFileSize int + mediaStore media.MediaStore + + defaultChannel string + defaultChatID string +} + +func NewSendFileTool(workspace string, restrict bool, maxFileSize int, store media.MediaStore) *SendFileTool { + if maxFileSize <= 0 { + maxFileSize = config.DefaultMaxMediaSize + } + return &SendFileTool{ + workspace: workspace, + restrict: restrict, + maxFileSize: maxFileSize, + mediaStore: store, + } +} + +func (t *SendFileTool) Name() string { return "send_file" } +func (t *SendFileTool) Description() string { + return "Send a local file (image, document, etc.) to the user on the current chat channel." +} + +func (t *SendFileTool) Parameters() map[string]any { + return map[string]any{ + "type": "object", + "properties": map[string]any{ + "path": map[string]any{ + "type": "string", + "description": "Path to the local file. Relative paths are resolved from workspace.", + }, + "filename": map[string]any{ + "type": "string", + "description": "Optional display filename. Defaults to the basename of path.", + }, + }, + "required": []string{"path"}, + } +} + +func (t *SendFileTool) SetContext(channel, chatID string) { + t.defaultChannel = channel + t.defaultChatID = chatID +} + +func (t *SendFileTool) SetMediaStore(store media.MediaStore) { + t.mediaStore = store +} + +func (t *SendFileTool) Execute(ctx context.Context, args map[string]any) *ToolResult { + path, _ := args["path"].(string) + if strings.TrimSpace(path) == "" { + return ErrorResult("path is required") + } + + // Prefer context-injected channel/chatID (set by ExecuteWithContext), fall back to SetContext values. + channel := ToolChannel(ctx) + if channel == "" { + channel = t.defaultChannel + } + chatID := ToolChatID(ctx) + if chatID == "" { + chatID = t.defaultChatID + } + if channel == "" || chatID == "" { + return ErrorResult("no target channel/chat available") + } + + if t.mediaStore == nil { + return ErrorResult("media store not configured") + } + + resolved, err := validatePath(path, t.workspace, t.restrict) + if err != nil { + return ErrorResult(fmt.Sprintf("invalid path: %v", err)) + } + + info, err := os.Stat(resolved) + if err != nil { + return ErrorResult(fmt.Sprintf("file not found: %v", err)) + } + if info.IsDir() { + return ErrorResult("path is a directory, expected a file") + } + if info.Size() > int64(t.maxFileSize) { + return ErrorResult(fmt.Sprintf( + "file too large: %d bytes (max %d bytes)", + info.Size(), t.maxFileSize, + )) + } + + filename, _ := args["filename"].(string) + if filename == "" { + filename = filepath.Base(resolved) + } + + mediaType := detectMediaType(resolved) + scope := fmt.Sprintf("tool:send_file:%s:%s", channel, chatID) + + ref, err := t.mediaStore.Store(resolved, media.MediaMeta{ + Filename: filename, + ContentType: mediaType, + Source: "tool:send_file", + }, scope) + if err != nil { + return ErrorResult(fmt.Sprintf("failed to register media: %v", err)) + } + + return MediaResult(fmt.Sprintf("File %q sent to user", filename), []string{ref}) +} + +// detectMediaType determines the MIME type of a file. +// Uses magic-bytes detection (h2non/filetype) first, then falls back to +// extension-based lookup via mime.TypeByExtension. +func detectMediaType(path string) string { + kind, err := filetype.MatchFile(path) + if err == nil && kind != filetype.Unknown { + return kind.MIME.Value + } + + if ext := filepath.Ext(path); ext != "" { + if t := mime.TypeByExtension(ext); t != "" { + return t + } + } + + return "application/octet-stream" +} diff --git a/pkg/tools/send_file_test.go b/pkg/tools/send_file_test.go new file mode 100644 index 000000000..08d129674 --- /dev/null +++ b/pkg/tools/send_file_test.go @@ -0,0 +1,176 @@ +package tools + +import ( + "context" + "os" + "path/filepath" + "strings" + "testing" + + "github.com/sipeed/picoclaw/pkg/config" + "github.com/sipeed/picoclaw/pkg/media" +) + +func TestSendFileTool_MissingPath(t *testing.T) { + store := media.NewFileMediaStore() + tool := NewSendFileTool("/tmp", false, 0, store) + tool.SetContext("feishu", "chat123") + + result := tool.Execute(context.Background(), map[string]any{}) + if !result.IsError { + t.Fatal("expected error for missing path") + } +} + +func TestSendFileTool_NoContext(t *testing.T) { + store := media.NewFileMediaStore() + tool := NewSendFileTool("/tmp", false, 0, store) + // no SetContext call + + result := tool.Execute(context.Background(), map[string]any{"path": "/tmp/test.txt"}) + if !result.IsError { + t.Fatal("expected error when no channel context") + } +} + +func TestSendFileTool_NoMediaStore(t *testing.T) { + tool := NewSendFileTool("/tmp", false, 0, nil) + tool.SetContext("feishu", "chat123") + + result := tool.Execute(context.Background(), map[string]any{"path": "/tmp/test.txt"}) + if !result.IsError { + t.Fatal("expected error when no media store") + } +} + +func TestSendFileTool_Directory(t *testing.T) { + store := media.NewFileMediaStore() + tool := NewSendFileTool("/tmp", false, 0, store) + tool.SetContext("feishu", "chat123") + + result := tool.Execute(context.Background(), map[string]any{"path": "/tmp"}) + if !result.IsError { + t.Fatal("expected error for directory path") + } +} + +func TestSendFileTool_FileTooLarge(t *testing.T) { + dir := t.TempDir() + testFile := filepath.Join(dir, "big.bin") + // Create a file larger than the limit + if err := os.WriteFile(testFile, make([]byte, 1024), 0o644); err != nil { + t.Fatal(err) + } + + store := media.NewFileMediaStore() + tool := NewSendFileTool(dir, false, 512, store) // 512 byte limit + tool.SetContext("feishu", "chat123") + + result := tool.Execute(context.Background(), map[string]any{"path": testFile}) + if !result.IsError { + t.Fatal("expected error for oversized file") + } + if !strings.Contains(result.ForLLM, "too large") { + t.Errorf("expected 'too large' in error, got %q", result.ForLLM) + } +} + +func TestSendFileTool_DefaultMaxSize(t *testing.T) { + tool := NewSendFileTool("/tmp", false, 0, nil) + if tool.maxFileSize != config.DefaultMaxMediaSize { + t.Errorf("expected default max size %d, got %d", config.DefaultMaxMediaSize, tool.maxFileSize) + } +} + +func TestSendFileTool_Success(t *testing.T) { + dir := t.TempDir() + testFile := filepath.Join(dir, "photo.png") + if err := os.WriteFile(testFile, []byte("fake png"), 0o644); err != nil { + t.Fatal(err) + } + + store := media.NewFileMediaStore() + tool := NewSendFileTool(dir, false, 0, store) + tool.SetContext("feishu", "chat123") + + result := tool.Execute(context.Background(), map[string]any{"path": testFile}) + if result.IsError { + t.Fatalf("unexpected error: %s", result.ForLLM) + } + if len(result.Media) != 1 { + t.Fatalf("expected 1 media ref, got %d", len(result.Media)) + } + if result.Media[0][:8] != "media://" { + t.Errorf("expected media:// ref, got %q", result.Media[0]) + } +} + +func TestSendFileTool_CustomFilename(t *testing.T) { + dir := t.TempDir() + testFile := filepath.Join(dir, "img.jpg") + if err := os.WriteFile(testFile, []byte("fake jpg"), 0o644); err != nil { + t.Fatal(err) + } + + store := media.NewFileMediaStore() + tool := NewSendFileTool(dir, false, 0, store) + tool.SetContext("telegram", "chat456") + + result := tool.Execute(context.Background(), map[string]any{ + "path": testFile, + "filename": "my-photo.jpg", + }) + if result.IsError { + t.Fatalf("unexpected error: %s", result.ForLLM) + } + if len(result.Media) != 1 { + t.Fatalf("expected 1 media ref, got %d", len(result.Media)) + } +} + +func TestDetectMediaType_MagicBytes(t *testing.T) { + dir := t.TempDir() + + // Minimal valid PNG header + pngHeader := []byte{0x89, 0x50, 0x4E, 0x47, 0x0D, 0x0A, 0x1A, 0x0A} + pngFile := filepath.Join(dir, "image.dat") // wrong extension, but valid PNG bytes + if err := os.WriteFile(pngFile, pngHeader, 0o644); err != nil { + t.Fatal(err) + } + + got := detectMediaType(pngFile) + if got != "image/png" { + t.Errorf("expected image/png from magic bytes, got %q", got) + } +} + +func TestDetectMediaType_FallbackToExtension(t *testing.T) { + dir := t.TempDir() + + // File with unrecognizable content but known extension + txtFile := filepath.Join(dir, "readme.txt") + if err := os.WriteFile(txtFile, []byte("hello world"), 0o644); err != nil { + t.Fatal(err) + } + + got := detectMediaType(txtFile) + // text/plain or similar — just verify it's not application/octet-stream + if got == "application/octet-stream" { + t.Errorf("expected extension-based MIME for .txt, got %q", got) + } +} + +func TestDetectMediaType_UnknownFallsToOctetStream(t *testing.T) { + dir := t.TempDir() + + // File with no extension and random bytes + unknownFile := filepath.Join(dir, "mystery") + if err := os.WriteFile(unknownFile, []byte{0x00, 0x01, 0x02}, 0o644); err != nil { + t.Fatal(err) + } + + got := detectMediaType(unknownFile) + if got != "application/octet-stream" { + t.Errorf("expected application/octet-stream, got %q", got) + } +} diff --git a/pkg/tools/shell.go b/pkg/tools/shell.go index a0c83eb1e..b8a811d03 100644 --- a/pkg/tools/shell.go +++ b/pkg/tools/shell.go @@ -59,7 +59,7 @@ var ( regexp.MustCompile(`\bchown\b`), regexp.MustCompile(`\bpkill\b`), regexp.MustCompile(`\bkillall\b`), - regexp.MustCompile(`\bkill\s+-[9]\b`), + regexp.MustCompile(`\bkill\b`), regexp.MustCompile(`\bcurl\b.*\|\s*(sh|bash)`), regexp.MustCompile(`\bwget\b.*\|\s*(sh|bash)`), regexp.MustCompile(`\bnpm\s+install\s+-g\b`), @@ -131,9 +131,14 @@ func NewExecToolWithConfig(workingDir string, restrict bool, config *config.Conf denyPatterns = append(denyPatterns, defaultDenyPatterns...) } + timeout := 60 * time.Second + if config != nil && config.Tools.Exec.TimeoutSeconds > 0 { + timeout = time.Duration(config.Tools.Exec.TimeoutSeconds) * time.Second + } + return &ExecTool{ workingDir: workingDir, - timeout: 60 * time.Second, + timeout: timeout, denyPatterns: denyPatterns, allowPatterns: nil, customAllowPatterns: customAllowPatterns, diff --git a/pkg/tools/shell_test.go b/pkg/tools/shell_test.go index a6abca8ea..ff9ea4a15 100644 --- a/pkg/tools/shell_test.go +++ b/pkg/tools/shell_test.go @@ -151,6 +151,26 @@ func TestShellTool_DangerousCommand(t *testing.T) { } } +func TestShellTool_DangerousCommand_KillBlocked(t *testing.T) { + tool, err := NewExecTool("", false) + if err != nil { + t.Errorf("unable to configure exec tool: %s", err) + } + + ctx := context.Background() + args := map[string]any{ + "command": "kill 12345", + } + + result := tool.Execute(ctx, args) + if !result.IsError { + t.Errorf("Expected kill command to be blocked") + } + if !strings.Contains(result.ForLLM, "blocked") && !strings.Contains(result.ForUser, "blocked") { + t.Errorf("Expected blocked message, got ForLLM: %s, ForUser: %s", result.ForLLM, result.ForUser) + } +} + // TestShellTool_MissingCommand verifies error handling for missing command func TestShellTool_MissingCommand(t *testing.T) { tool, err := NewExecTool("", false) diff --git a/pkg/tools/spawn.go b/pkg/tools/spawn.go index 8b166b41f..be40ffda2 100644 --- a/pkg/tools/spawn.go +++ b/pkg/tools/spawn.go @@ -8,25 +8,18 @@ import ( type SpawnTool struct { manager *SubagentManager - originChannel string - originChatID string allowlistCheck func(targetAgentID string) bool - callback AsyncCallback // For async completion notification } +// Compile-time check: SpawnTool implements AsyncExecutor. +var _ AsyncExecutor = (*SpawnTool)(nil) + func NewSpawnTool(manager *SubagentManager) *SpawnTool { return &SpawnTool{ - manager: manager, - originChannel: "cli", - originChatID: "direct", + manager: manager, } } -// SetCallback implements AsyncTool interface for async completion notification -func (t *SpawnTool) SetCallback(cb AsyncCallback) { - t.callback = cb -} - func (t *SpawnTool) Name() string { return "spawn" } @@ -56,16 +49,21 @@ func (t *SpawnTool) Parameters() map[string]any { } } -func (t *SpawnTool) SetContext(channel, chatID string) { - t.originChannel = channel - t.originChatID = chatID -} - func (t *SpawnTool) SetAllowlistChecker(check func(targetAgentID string) bool) { t.allowlistCheck = check } func (t *SpawnTool) Execute(ctx context.Context, args map[string]any) *ToolResult { + return t.execute(ctx, args, nil) +} + +// ExecuteAsync implements AsyncExecutor. The callback is passed through to the +// subagent manager as a call parameter — never stored on the SpawnTool instance. +func (t *SpawnTool) ExecuteAsync(ctx context.Context, args map[string]any, cb AsyncCallback) *ToolResult { + return t.execute(ctx, args, cb) +} + +func (t *SpawnTool) execute(ctx context.Context, args map[string]any, cb AsyncCallback) *ToolResult { task, ok := args["task"].(string) if !ok || strings.TrimSpace(task) == "" { return ErrorResult("task is required and must be a non-empty string") @@ -85,8 +83,20 @@ func (t *SpawnTool) Execute(ctx context.Context, args map[string]any) *ToolResul return ErrorResult("Subagent manager not configured") } + // Read channel/chatID from context (injected by registry). + // Fall back to "cli"/"direct" for non-conversation callers (e.g., CLI, tests) + // to preserve the same defaults as the original NewSpawnTool constructor. + channel := ToolChannel(ctx) + if channel == "" { + channel = "cli" + } + chatID := ToolChatID(ctx) + if chatID == "" { + chatID = "direct" + } + // Pass callback to manager for async completion notification - result, err := t.manager.Spawn(ctx, task, label, agentID, t.originChannel, t.originChatID, t.callback) + result, err := t.manager.Spawn(ctx, task, label, agentID, channel, chatID, cb) if err != nil { return ErrorResult(fmt.Sprintf("failed to spawn subagent: %v", err)) } diff --git a/pkg/tools/subagent.go b/pkg/tools/subagent.go index 69f1a49a2..429340047 100644 --- a/pkg/tools/subagent.go +++ b/pkg/tools/subagent.go @@ -252,16 +252,12 @@ func (sm *SubagentManager) ListTasks() []*SubagentTask { // Unlike SpawnTool which runs tasks asynchronously, SubagentTool waits for completion // and returns the result directly in the ToolResult. type SubagentTool struct { - manager *SubagentManager - originChannel string - originChatID string + manager *SubagentManager } func NewSubagentTool(manager *SubagentManager) *SubagentTool { return &SubagentTool{ - manager: manager, - originChannel: "cli", - originChatID: "direct", + manager: manager, } } @@ -290,11 +286,6 @@ func (t *SubagentTool) Parameters() map[string]any { } } -func (t *SubagentTool) SetContext(channel, chatID string) { - t.originChannel = channel - t.originChatID = chatID -} - func (t *SubagentTool) Execute(ctx context.Context, args map[string]any) *ToolResult { task, ok := args["task"].(string) if !ok { @@ -341,13 +332,24 @@ func (t *SubagentTool) Execute(ctx context.Context, args map[string]any) *ToolRe } } + // Fall back to "cli"/"direct" for non-conversation callers (e.g., CLI, tests) + // to preserve the same defaults as the original NewSubagentTool constructor. + channel := ToolChannel(ctx) + if channel == "" { + channel = "cli" + } + chatID := ToolChatID(ctx) + if chatID == "" { + chatID = "direct" + } + loopResult, err := RunToolLoop(ctx, ToolLoopConfig{ Provider: sm.provider, Model: sm.defaultModel, Tools: tools, MaxIterations: maxIter, LLMOptions: llmOptions, - }, messages, t.originChannel, t.originChatID) + }, messages, channel, chatID) if err != nil { return ErrorResult(fmt.Sprintf("Subagent execution failed: %v", err)).WithError(err) } diff --git a/pkg/tools/subagent_tool_test.go b/pkg/tools/subagent_tool_test.go index 59bfdffae..a1450410a 100644 --- a/pkg/tools/subagent_tool_test.go +++ b/pkg/tools/subagent_tool_test.go @@ -50,9 +50,8 @@ func TestSubagentManager_SetLLMOptions_AppliesToRunToolLoop(t *testing.T) { manager := NewSubagentManager(provider, "test-model", "/tmp/test", nil) manager.SetLLMOptions(2048, 0.6) tool := NewSubagentTool(manager) - tool.SetContext("cli", "direct") - ctx := context.Background() + ctx := WithToolContext(context.Background(), "cli", "direct") args := map[string]any{"task": "Do something"} result := tool.Execute(ctx, args) @@ -147,28 +146,14 @@ func TestSubagentTool_Parameters(t *testing.T) { } } -// TestSubagentTool_SetContext verifies context setting -func TestSubagentTool_SetContext(t *testing.T) { - provider := &MockLLMProvider{} - manager := NewSubagentManager(provider, "test-model", "/tmp/test", nil) - tool := NewSubagentTool(manager) - - tool.SetContext("test-channel", "test-chat") - - // Verify context is set (we can't directly access private fields, - // but we can verify it doesn't crash) - // The actual context usage is tested in Execute tests -} - // TestSubagentTool_Execute_Success tests successful execution func TestSubagentTool_Execute_Success(t *testing.T) { provider := &MockLLMProvider{} msgBus := bus.NewMessageBus() manager := NewSubagentManager(provider, "test-model", "/tmp/test", msgBus) tool := NewSubagentTool(manager) - tool.SetContext("telegram", "chat-123") - ctx := context.Background() + ctx := WithToolContext(context.Background(), "telegram", "chat-123") args := map[string]any{ "task": "Write a haiku about coding", "label": "haiku-task", @@ -297,12 +282,9 @@ func TestSubagentTool_Execute_ContextPassing(t *testing.T) { manager := NewSubagentManager(provider, "test-model", "/tmp/test", msgBus) tool := NewSubagentTool(manager) - // Set context channel := "test-channel" chatID := "test-chat" - tool.SetContext(channel, chatID) - - ctx := context.Background() + ctx := WithToolContext(context.Background(), channel, chatID) args := map[string]any{ "task": "Test context passing", } diff --git a/pkg/tools/toolloop.go b/pkg/tools/toolloop.go index cdfe0d6ce..244f0d4a2 100644 --- a/pkg/tools/toolloop.go +++ b/pkg/tools/toolloop.go @@ -10,6 +10,7 @@ import ( "context" "encoding/json" "fmt" + "sync" "github.com/sipeed/picoclaw/pkg/logger" "github.com/sipeed/picoclaw/pkg/providers" @@ -121,37 +122,53 @@ func RunToolLoop( } messages = append(messages, assistantMsg) - // 7. Execute tool calls - for _, tc := range normalizedToolCalls { - argsJSON, _ := json.Marshal(tc.Arguments) - argsPreview := utils.Truncate(string(argsJSON), 200) - logger.InfoCF("toolloop", fmt.Sprintf("Tool call: %s(%s)", tc.Name, argsPreview), - map[string]any{ - "tool": tc.Name, - "iteration": iteration, - }) + // 7. Execute tool calls in parallel + type indexedResult struct { + result *ToolResult + tc providers.ToolCall + } - // Execute tool (no async callback for subagents - they run independently) - var toolResult *ToolResult - if config.Tools != nil { - toolResult = config.Tools.ExecuteWithContext(ctx, tc.Name, tc.Arguments, channel, chatID, nil) - } else { - toolResult = ErrorResult("No tools available") + results := make([]indexedResult, len(normalizedToolCalls)) + var wg sync.WaitGroup + + for i, tc := range normalizedToolCalls { + results[i].tc = tc + + wg.Add(1) + go func(idx int, tc providers.ToolCall) { + defer wg.Done() + + argsJSON, _ := json.Marshal(tc.Arguments) + argsPreview := utils.Truncate(string(argsJSON), 200) + logger.InfoCF("toolloop", fmt.Sprintf("Tool call: %s(%s)", tc.Name, argsPreview), + map[string]any{ + "tool": tc.Name, + "iteration": iteration, + }) + + var toolResult *ToolResult + if config.Tools != nil { + toolResult = config.Tools.ExecuteWithContext(ctx, tc.Name, tc.Arguments, channel, chatID, nil) + } else { + toolResult = ErrorResult("No tools available") + } + results[idx].result = toolResult + }(i, tc) + } + wg.Wait() + + // Append results in original order + for _, r := range results { + contentForLLM := r.result.ForLLM + if contentForLLM == "" && r.result.Err != nil { + contentForLLM = r.result.Err.Error() } - // Determine content for LLM - contentForLLM := toolResult.ForLLM - if contentForLLM == "" && toolResult.Err != nil { - contentForLLM = toolResult.Err.Error() - } - - // Add tool result message - toolResultMsg := providers.Message{ + messages = append(messages, providers.Message{ Role: "tool", Content: contentForLLM, - ToolCallID: tc.ID, - } - messages = append(messages, toolResultMsg) + ToolCallID: r.tc.ID, + }) } } diff --git a/pkg/tools/web.go b/pkg/tools/web.go index 15d2330ff..eeceabd98 100644 --- a/pkg/tools/web.go +++ b/pkg/tools/web.go @@ -395,6 +395,150 @@ func (p *PerplexitySearchProvider) Search(ctx context.Context, query string, cou return fmt.Sprintf("Results for: %s (via Perplexity)\n%s", query, searchResp.Choices[0].Message.Content), nil } +type SearXNGSearchProvider struct { + baseURL string +} + +func (p *SearXNGSearchProvider) Search(ctx context.Context, query string, count int) (string, error) { + searchURL := fmt.Sprintf("%s/search?q=%s&format=json&categories=general", + strings.TrimSuffix(p.baseURL, "/"), + url.QueryEscape(query)) + + req, err := http.NewRequestWithContext(ctx, "GET", searchURL, nil) + if err != nil { + return "", fmt.Errorf("failed to create request: %w", err) + } + + client := &http.Client{Timeout: 10 * time.Second} + resp, err := client.Do(req) + if err != nil { + return "", fmt.Errorf("request failed: %w", err) + } + defer resp.Body.Close() + + if resp.StatusCode != http.StatusOK { + return "", fmt.Errorf("SearXNG returned status %d", resp.StatusCode) + } + + var result struct { + Results []struct { + Title string `json:"title"` + URL string `json:"url"` + Content string `json:"content"` + Engine string `json:"engine"` + Score float64 `json:"score"` + } `json:"results"` + } + + if err := json.NewDecoder(resp.Body).Decode(&result); err != nil { + return "", fmt.Errorf("failed to parse response: %w", err) + } + + if len(result.Results) == 0 { + return fmt.Sprintf("No results for: %s", query), nil + } + + // Limit results to requested count + if len(result.Results) > count { + result.Results = result.Results[:count] + } + + // Format results in standard PicoClaw format + var b strings.Builder + b.WriteString(fmt.Sprintf("Results for: %s (via SearXNG)\n", query)) + for i, r := range result.Results { + b.WriteString(fmt.Sprintf("%d. %s\n", i+1, r.Title)) + b.WriteString(fmt.Sprintf(" %s\n", r.URL)) + if r.Content != "" { + b.WriteString(fmt.Sprintf(" %s\n", r.Content)) + } + } + + return b.String(), nil +} + +type GLMSearchProvider struct { + apiKey string + baseURL string + searchEngine string + proxy string + client *http.Client +} + +func (p *GLMSearchProvider) Search(ctx context.Context, query string, count int) (string, error) { + searchURL := p.baseURL + if searchURL == "" { + searchURL = "https://open.bigmodel.cn/api/paas/v4/web_search" + } + + payload := map[string]any{ + "search_query": query, + "search_engine": p.searchEngine, + "search_intent": false, + "count": count, + "content_size": "medium", + } + + bodyBytes, err := json.Marshal(payload) + if err != nil { + return "", fmt.Errorf("failed to marshal payload: %w", err) + } + + req, err := http.NewRequestWithContext(ctx, "POST", searchURL, bytes.NewReader(bodyBytes)) + if err != nil { + return "", fmt.Errorf("failed to create request: %w", err) + } + + req.Header.Set("Content-Type", "application/json") + req.Header.Set("Authorization", "Bearer "+p.apiKey) + + resp, err := p.client.Do(req) + if err != nil { + return "", fmt.Errorf("request failed: %w", err) + } + defer resp.Body.Close() + + body, err := io.ReadAll(io.LimitReader(resp.Body, 1<<20)) + if err != nil { + return "", fmt.Errorf("failed to read response: %w", err) + } + + if resp.StatusCode != http.StatusOK { + return "", fmt.Errorf("GLM Search API error (status %d): %s", resp.StatusCode, string(body)) + } + + var searchResp struct { + SearchResult []struct { + Title string `json:"title"` + Content string `json:"content"` + Link string `json:"link"` + } `json:"search_result"` + } + + if err := json.Unmarshal(body, &searchResp); err != nil { + return "", fmt.Errorf("failed to parse response: %w", err) + } + + results := searchResp.SearchResult + if len(results) == 0 { + return fmt.Sprintf("No results for: %s", query), nil + } + + var lines []string + lines = append(lines, fmt.Sprintf("Results for: %s (via GLM Search)", query)) + for i, item := range results { + if i >= count { + break + } + lines = append(lines, fmt.Sprintf("%d. %s\n %s", i+1, item.Title, item.Link)) + if item.Content != "" { + lines = append(lines, fmt.Sprintf(" %s", item.Content)) + } + } + + return strings.Join(lines, "\n"), nil +} + type WebSearchTool struct { provider SearchProvider maxResults int @@ -413,6 +557,14 @@ type WebSearchToolOptions struct { PerplexityAPIKey string PerplexityMaxResults int PerplexityEnabled bool + SearXNGBaseURL string + SearXNGMaxResults int + SearXNGEnabled bool + GLMSearchAPIKey string + GLMSearchBaseURL string + GLMSearchEngine string + GLMSearchMaxResults int + GLMSearchEnabled bool Proxy string } @@ -420,7 +572,7 @@ func NewWebSearchTool(opts WebSearchToolOptions) (*WebSearchTool, error) { var provider SearchProvider maxResults := 5 - // Priority: Perplexity > Brave > Tavily > DuckDuckGo + // Priority: Perplexity > Brave > SearXNG > Tavily > DuckDuckGo > GLM Search if opts.PerplexityEnabled && opts.PerplexityAPIKey != "" { client, err := createHTTPClient(opts.Proxy, perplexityTimeout) if err != nil { @@ -439,6 +591,11 @@ func NewWebSearchTool(opts WebSearchToolOptions) (*WebSearchTool, error) { if opts.BraveMaxResults > 0 { maxResults = opts.BraveMaxResults } + } else if opts.SearXNGEnabled && opts.SearXNGBaseURL != "" { + provider = &SearXNGSearchProvider{baseURL: opts.SearXNGBaseURL} + if opts.SearXNGMaxResults > 0 { + maxResults = opts.SearXNGMaxResults + } } else if opts.TavilyEnabled && opts.TavilyAPIKey != "" { client, err := createHTTPClient(opts.Proxy, searchTimeout) if err != nil { @@ -462,6 +619,25 @@ func NewWebSearchTool(opts WebSearchToolOptions) (*WebSearchTool, error) { if opts.DuckDuckGoMaxResults > 0 { maxResults = opts.DuckDuckGoMaxResults } + } else if opts.GLMSearchEnabled && opts.GLMSearchAPIKey != "" { + client, err := createHTTPClient(opts.Proxy, searchTimeout) + if err != nil { + return nil, fmt.Errorf("failed to create HTTP client for GLM Search: %w", err) + } + searchEngine := opts.GLMSearchEngine + if searchEngine == "" { + searchEngine = "search_std" + } + provider = &GLMSearchProvider{ + apiKey: opts.GLMSearchAPIKey, + baseURL: opts.GLMSearchBaseURL, + searchEngine: searchEngine, + proxy: opts.Proxy, + client: client, + } + if opts.GLMSearchMaxResults > 0 { + maxResults = opts.GLMSearchMaxResults + } } else { return nil, nil } diff --git a/pkg/tools/web_test.go b/pkg/tools/web_test.go index 8a8b88131..bdd30d385 100644 --- a/pkg/tools/web_test.go +++ b/pkg/tools/web_test.go @@ -681,3 +681,135 @@ func TestWebTool_TavilySearch_Success(t *testing.T) { t.Errorf("Expected 'via Tavily' in output, got: %s", result.ForUser) } } + +func TestWebTool_GLMSearch_Success(t *testing.T) { + server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + if r.Method != "POST" { + t.Errorf("Expected POST request, got %s", r.Method) + } + if r.Header.Get("Content-Type") != "application/json" { + t.Errorf("Expected Content-Type application/json, got %s", r.Header.Get("Content-Type")) + } + if r.Header.Get("Authorization") != "Bearer test-glm-key" { + t.Errorf("Expected Authorization Bearer test-glm-key, got %s", r.Header.Get("Authorization")) + } + + var payload map[string]any + json.NewDecoder(r.Body).Decode(&payload) + if payload["search_query"] != "test query" { + t.Errorf("Expected search_query 'test query', got %v", payload["search_query"]) + } + if payload["search_engine"] != "search_std" { + t.Errorf("Expected search_engine 'search_std', got %v", payload["search_engine"]) + } + + response := map[string]any{ + "id": "web-search-test", + "created": 1709568000, + "search_result": []map[string]any{ + { + "title": "Test GLM Result", + "content": "GLM search snippet", + "link": "https://example.com/glm", + "media": "Example", + "publish_date": "2026-03-04", + }, + }, + } + w.Header().Set("Content-Type", "application/json") + w.WriteHeader(http.StatusOK) + json.NewEncoder(w).Encode(response) + })) + defer server.Close() + + tool, err := NewWebSearchTool(WebSearchToolOptions{ + GLMSearchEnabled: true, + GLMSearchAPIKey: "test-glm-key", + GLMSearchBaseURL: server.URL, + GLMSearchEngine: "search_std", + }) + if err != nil { + t.Fatalf("NewWebSearchTool() error: %v", err) + } + + result := tool.Execute(context.Background(), map[string]any{ + "query": "test query", + }) + + if result.IsError { + t.Errorf("Expected success, got IsError=true: %s", result.ForLLM) + } + if !strings.Contains(result.ForUser, "Test GLM Result") { + t.Errorf("Expected 'Test GLM Result' in output, got: %s", result.ForUser) + } + if !strings.Contains(result.ForUser, "https://example.com/glm") { + t.Errorf("Expected URL in output, got: %s", result.ForUser) + } + if !strings.Contains(result.ForUser, "via GLM Search") { + t.Errorf("Expected 'via GLM Search' in output, got: %s", result.ForUser) + } +} + +func TestWebTool_GLMSearch_APIError(t *testing.T) { + server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, _ *http.Request) { + w.WriteHeader(http.StatusUnauthorized) + w.Write([]byte(`{"error":"invalid api key"}`)) + })) + defer server.Close() + + tool, err := NewWebSearchTool(WebSearchToolOptions{ + GLMSearchEnabled: true, + GLMSearchAPIKey: "bad-key", + GLMSearchBaseURL: server.URL, + GLMSearchEngine: "search_std", + }) + if err != nil { + t.Fatalf("NewWebSearchTool() error: %v", err) + } + + result := tool.Execute(context.Background(), map[string]any{ + "query": "test query", + }) + + if !result.IsError { + t.Errorf("Expected IsError=true for 401 response") + } + if !strings.Contains(result.ForLLM, "status 401") { + t.Errorf("Expected status 401 in error, got: %s", result.ForLLM) + } +} + +func TestWebTool_GLMSearch_Priority(t *testing.T) { + // GLM Search should only be selected when all other providers are disabled + tool, err := NewWebSearchTool(WebSearchToolOptions{ + DuckDuckGoEnabled: true, + DuckDuckGoMaxResults: 5, + GLMSearchEnabled: true, + GLMSearchAPIKey: "test-key", + GLMSearchBaseURL: "https://example.com", + GLMSearchEngine: "search_std", + }) + if err != nil { + t.Fatalf("NewWebSearchTool() error: %v", err) + } + + // DuckDuckGo should win over GLM Search + if _, ok := tool.provider.(*DuckDuckGoSearchProvider); !ok { + t.Errorf("Expected DuckDuckGoSearchProvider when both enabled, got %T", tool.provider) + } + + // With DuckDuckGo disabled, GLM Search should be selected + tool2, err := NewWebSearchTool(WebSearchToolOptions{ + DuckDuckGoEnabled: false, + GLMSearchEnabled: true, + GLMSearchAPIKey: "test-key", + GLMSearchBaseURL: "https://example.com", + GLMSearchEngine: "search_std", + }) + if err != nil { + t.Fatalf("NewWebSearchTool() error: %v", err) + } + if _, ok := tool2.provider.(*GLMSearchProvider); !ok { + t.Errorf("Expected GLMSearchProvider when only GLM enabled, got %T", tool2.provider) + } +} diff --git a/pkg/utils/media.go b/pkg/utils/media.go index a34889fb8..3e1c5d88e 100644 --- a/pkg/utils/media.go +++ b/pkg/utils/media.go @@ -3,6 +3,7 @@ package utils import ( "io" "net/http" + "net/url" "os" "path/filepath" "strings" @@ -52,11 +53,12 @@ type DownloadOptions struct { Timeout time.Duration ExtraHeaders map[string]string LoggerPrefix string + ProxyURL string } // DownloadFile downloads a file from URL to a local temp directory. // Returns the local file path or empty string on error. -func DownloadFile(url, filename string, opts DownloadOptions) string { +func DownloadFile(urlStr, filename string, opts DownloadOptions) string { // Set defaults if opts.Timeout == 0 { opts.Timeout = 60 * time.Second @@ -78,7 +80,7 @@ func DownloadFile(url, filename string, opts DownloadOptions) string { localPath := filepath.Join(mediaDir, uuid.New().String()[:8]+"_"+safeName) // Create HTTP request - req, err := http.NewRequest("GET", url, nil) + req, err := http.NewRequest("GET", urlStr, nil) if err != nil { logger.ErrorCF(opts.LoggerPrefix, "Failed to create download request", map[string]any{ "error": err.Error(), @@ -92,11 +94,24 @@ func DownloadFile(url, filename string, opts DownloadOptions) string { } client := &http.Client{Timeout: opts.Timeout} + if opts.ProxyURL != "" { + proxyURL, parseErr := url.Parse(opts.ProxyURL) + if parseErr != nil { + logger.ErrorCF(opts.LoggerPrefix, "Invalid proxy URL for download", map[string]any{ + "error": parseErr.Error(), + "proxy": opts.ProxyURL, + }) + return "" + } + client.Transport = &http.Transport{ + Proxy: http.ProxyURL(proxyURL), + } + } resp, err := client.Do(req) if err != nil { logger.ErrorCF(opts.LoggerPrefix, "Failed to download file", map[string]any{ "error": err.Error(), - "url": url, + "url": urlStr, }) return "" } @@ -105,7 +120,7 @@ func DownloadFile(url, filename string, opts DownloadOptions) string { if resp.StatusCode != http.StatusOK { logger.ErrorCF(opts.LoggerPrefix, "File download returned non-200 status", map[string]any{ "status": resp.StatusCode, - "url": url, + "url": urlStr, }) return "" } diff --git a/pkg/voice/transcriber.go b/pkg/voice/transcriber.go index f973e77fe..e949d7a22 100644 --- a/pkg/voice/transcriber.go +++ b/pkg/voice/transcriber.go @@ -10,12 +10,19 @@ import ( "net/http" "os" "path/filepath" + "strings" "time" + "github.com/sipeed/picoclaw/pkg/config" "github.com/sipeed/picoclaw/pkg/logger" "github.com/sipeed/picoclaw/pkg/utils" ) +type Transcriber interface { + Name() string + Transcribe(ctx context.Context, audioFilePath string) (*TranscriptionResponse, error) +} + type GroqTranscriber struct { apiKey string apiBase string @@ -152,8 +159,22 @@ func (t *GroqTranscriber) Transcribe(ctx context.Context, audioFilePath string) return &result, nil } -func (t *GroqTranscriber) IsAvailable() bool { - available := t.apiKey != "" - logger.DebugCF("voice", "Checking transcriber availability", map[string]any{"available": available}) - return available +func (t *GroqTranscriber) Name() string { + return "groq" +} + +// DetectTranscriber inspects cfg and returns the appropriate Transcriber, or +// nil if no supported transcription provider is configured. +func DetectTranscriber(cfg *config.Config) Transcriber { + // Direct Groq provider config takes priority. + if key := cfg.Providers.Groq.APIKey; key != "" { + return NewGroqTranscriber(key) + } + // Fall back to any model-list entry that uses the groq/ protocol. + for _, mc := range cfg.ModelList { + if strings.HasPrefix(mc.Model, "groq/") && mc.APIKey != "" { + return NewGroqTranscriber(mc.APIKey) + } + } + return nil } diff --git a/pkg/voice/transcriber_test.go b/pkg/voice/transcriber_test.go new file mode 100644 index 000000000..9b6add333 --- /dev/null +++ b/pkg/voice/transcriber_test.go @@ -0,0 +1,160 @@ +package voice + +import ( + "context" + "encoding/json" + "net/http" + "net/http/httptest" + "os" + "path/filepath" + "testing" + + "github.com/sipeed/picoclaw/pkg/config" +) + +// Ensure GroqTranscriber satisfies the Transcriber interface at compile time. +var _ Transcriber = (*GroqTranscriber)(nil) + +func TestGroqTranscriberName(t *testing.T) { + tr := NewGroqTranscriber("sk-test") + if got := tr.Name(); got != "groq" { + t.Errorf("Name() = %q, want %q", got, "groq") + } +} + +func TestDetectTranscriber(t *testing.T) { + tests := []struct { + name string + cfg *config.Config + wantNil bool + wantName string + }{ + { + name: "no config", + cfg: &config.Config{}, + wantNil: true, + }, + { + name: "groq provider key", + cfg: &config.Config{ + Providers: config.ProvidersConfig{ + Groq: config.ProviderConfig{APIKey: "sk-groq-direct"}, + }, + }, + wantName: "groq", + }, + { + name: "groq via model list", + cfg: &config.Config{ + ModelList: []config.ModelConfig{ + {Model: "openai/gpt-4o", APIKey: "sk-openai"}, + {Model: "groq/llama-3.3-70b", APIKey: "sk-groq-model"}, + }, + }, + wantName: "groq", + }, + { + name: "groq model list entry without key is skipped", + cfg: &config.Config{ + ModelList: []config.ModelConfig{ + {Model: "groq/llama-3.3-70b", APIKey: ""}, + }, + }, + wantNil: true, + }, + { + name: "provider key takes priority over model list", + cfg: &config.Config{ + Providers: config.ProvidersConfig{ + Groq: config.ProviderConfig{APIKey: "sk-groq-direct"}, + }, + ModelList: []config.ModelConfig{ + {Model: "groq/llama-3.3-70b", APIKey: "sk-groq-model"}, + }, + }, + wantName: "groq", + }, + } + + for _, tc := range tests { + t.Run(tc.name, func(t *testing.T) { + tr := DetectTranscriber(tc.cfg) + if tc.wantNil { + if tr != nil { + t.Errorf("DetectTranscriber() = %v, want nil", tr) + } + return + } + if tr == nil { + t.Fatal("DetectTranscriber() = nil, want non-nil") + } + if got := tr.Name(); got != tc.wantName { + t.Errorf("Name() = %q, want %q", got, tc.wantName) + } + }) + } +} + +func TestTranscribe(t *testing.T) { + // Write a minimal fake audio file so the transcriber can open and send it. + tmpDir := t.TempDir() + audioPath := filepath.Join(tmpDir, "clip.ogg") + if err := os.WriteFile(audioPath, []byte("fake-audio-data"), 0o644); err != nil { + t.Fatalf("failed to write fake audio file: %v", err) + } + + t.Run("success", func(t *testing.T) { + srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + if r.URL.Path != "/audio/transcriptions" { + t.Errorf("unexpected path: %s", r.URL.Path) + } + if r.Header.Get("Authorization") != "Bearer sk-test" { + t.Errorf("unexpected Authorization header: %s", r.Header.Get("Authorization")) + } + w.Header().Set("Content-Type", "application/json") + _ = json.NewEncoder(w).Encode(TranscriptionResponse{ + Text: "hello world", + Language: "en", + Duration: 1.5, + }) + })) + defer srv.Close() + + tr := NewGroqTranscriber("sk-test") + tr.apiBase = srv.URL + + resp, err := tr.Transcribe(context.Background(), audioPath) + if err != nil { + t.Fatalf("Transcribe() error: %v", err) + } + if resp.Text != "hello world" { + t.Errorf("Text = %q, want %q", resp.Text, "hello world") + } + if resp.Language != "en" { + t.Errorf("Language = %q, want %q", resp.Language, "en") + } + }) + + t.Run("api error", func(t *testing.T) { + srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + http.Error(w, `{"error":"invalid_api_key"}`, http.StatusUnauthorized) + })) + defer srv.Close() + + tr := NewGroqTranscriber("sk-bad") + tr.apiBase = srv.URL + + _, err := tr.Transcribe(context.Background(), audioPath) + if err == nil { + t.Fatal("expected error for non-200 response, got nil") + } + }) + + t.Run("missing file", func(t *testing.T) { + tr := NewGroqTranscriber("sk-test") + _, err := tr.Transcribe(context.Background(), filepath.Join(tmpDir, "nonexistent.ogg")) + if err == nil { + t.Fatal("expected error for missing file, got nil") + } + }) +}