opencode/picoclaw
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

feat(agent): wire Media through agent pipeline (cherry-pick PR #555)

Browse source 
Add Media field to processOptions, pass msg.Media from inbound
messages through to BuildMessages and serializeMessages so
vision-capable LLMs receive image_url content parts.

Based on work by @as3k in sipeed/picoclaw#555.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
shikihane 2026-03-01 19:19:31 +08:00
parent 3d54a77c40
commit 6997edc82e
2 changed files with 10 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
pkg/agent/context.go
View file

@ -465,10 +465,11 @@ func (cb *ContextBuilder) BuildMessages(
messages = append(messages, history...) messages = append(messages, history...)
// Add current user message // Add current user message
if strings.TrimSpace(currentMessage) != "" { if strings.TrimSpace(currentMessage) != "" || len(media) > 0 {
messages = append(messages, providers.Message{ messages = append(messages, providers.Message{
Role: "user", Role: "user",
Content: currentMessage, Content: currentMessage,
Media: media,
}) })
} }

14
pkg/agent/loop.go
View file

@ -46,11 +46,12 @@ type AgentLoop struct {
// processOptions configures how a message is processed // processOptions configures how a message is processed
type processOptions struct { type processOptions struct {
SessionKey string // Session identifier for history/context SessionKey string // Session identifier for history/context
Channel string // Target channel for tool execution Channel string // Target channel for tool execution
ChatID string // Target chat ID for tool execution ChatID string // Target chat ID for tool execution
UserMessage string // User message content (may include prefix) UserMessage string // User message content (may include prefix)
DefaultResponse string // Response when LLM returns empty Media []string // Media URLs attached to the user message
DefaultResponse string // Response when LLM returns empty
EnableSummary bool // Whether to trigger summarization EnableSummary bool // Whether to trigger summarization
SendResponse bool // Whether to send response via bus SendResponse bool // Whether to send response via bus
NoHistory bool // If true, don't load session history (for heartbeat) NoHistory bool // If true, don't load session history (for heartbeat)
@ -417,6 +418,7 @@ func (al *AgentLoop) processMessage(ctx context.Context, msg bus.InboundMessage)
Channel: msg.Channel, Channel: msg.Channel,
ChatID: msg.ChatID, ChatID: msg.ChatID,
UserMessage: msg.Content, UserMessage: msg.Content,
Media: msg.Media,
DefaultResponse: defaultResponse, DefaultResponse: defaultResponse,
EnableSummary: true, EnableSummary: true,
SendResponse: false, SendResponse: false,
@ -509,7 +511,7 @@ func (al *AgentLoop) runAgentLoop(ctx context.Context, agent *AgentInstance, opt
history, history,
summary, summary,
opts.UserMessage, opts.UserMessage,
nil, opts.Media,
opts.Channel, opts.Channel,
opts.ChatID, opts.ChatID,
) )