feat: 增强 auto mode 的易用性 (#312)

* feat: poor 模式降级 yolo 审阅模型

* feat: 为多模块添加 Langfuse tracing 支持

在 web search、agent creation、away summary、token estimation、
skill improvement 等模块中集成 Langfuse trace,并透传至
compact/apiQueryHook/execPromptHook 等调用链。

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: 让 auto mode 记录回主 trace

* fix: reopen auto mode prompt when classifier is unavailable

* fix: 修复 auto mode 情况下, llm 报错导致弹窗也不打开的问题

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
claude-code-best
2026-04-20 21:13:09 +08:00
committed by GitHub
parent e4ce08fe39
commit ed4bdb9338
18 changed files with 281 additions and 145 deletions

View File

@@ -277,6 +277,8 @@ export type ToolUseContext = {
criticalSystemReminder_EXPERIMENTAL?: string
/** Langfuse root trace span for this query turn. Passed down to tool execution for observability. */
langfuseTrace?: LangfuseSpan | null
/** Langfuse root trace span for the outer/main agent trace. Used when subagents need to nest observations under the parent agent trace. */
langfuseRootTrace?: LangfuseSpan | null
/** Langfuse batch span wrapping a concurrent tool group. When set, tool observations are nested under it. */
langfuseBatchSpan?: LangfuseSpan | null
/** When true, preserve toolUseResult on messages even for subagents.