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

fix: preserve ReasoningContent in toolloop for thinking models

Browse source 
When using thinking models (e.g., Kimi K2.5), the API requires
all assistant messages with tool calls to include reasoning_content.
This field was missing in RunToolLoop, causing subagent failures with
the error: 已启用思考功能，但索引2处的助手工具调用消息中缺少推理内容

Aligns toolloop.go with the existing handling in agent/loop.go.
This commit is contained in:
mxrain 2026-02-15 11:38:28 +08:00
parent 39998a6618
commit 53ddf35b78
1 changed files with 3 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
pkg/tools/toolloop.go
View file

@ -99,6 +99,7 @@ func RunToolLoop(ctx context.Context, config ToolLoopConfig, messages []provider
assistantMsg := providers.Message{ assistantMsg := providers.Message{
Role: "assistant", Role: "assistant",
Content: response.Content, Content: response.Content,
ReasoningContent: response.ReasoningContent, // Preserve for thinking models (e.g., GLM-Z1)
} }
for _, tc := range response.ToolCalls { for _, tc := range response.ToolCalls {
argumentsJSON, _ := json.Marshal(tc.Arguments) argumentsJSON, _ := json.Marshal(tc.Arguments)