xiaohei/Agentswarm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Feat/queen convergence #24

Merged
xiaohei merged 2 commits from feat/queen-convergence into main 2026-06-21 20:00:36 +00:00
Conversation 0 Commits 2 Files Changed 1
+30 -8
xiaohei commented 2026-06-21 20:00:32 +00:00
Owner
Copy Link
Copy Source

变更说明

影响范围

  • Client
  • Manager
  • Agent / Swarm
  • CodeGW
  • 计费
  • 密钥 / secret_ref
  • 审计
  • 发布链路
  • 文档 / 标准

是否读取标准

  • 已读取当前仓 CLAUDE.md
  • 已读取当前仓 PROJECT_STANDARD.md
  • 已读取 heicodeDocs 相关标准

是否涉及接口契约

  • 不涉及
  • 涉及，已更新 docs/integration 或相关文档

验收方式

TODO / Mock

  • 没有 TODO / mock
  • 有，说明原因和后续计划

Release 仓库确认

  • 本 PR 不涉及 release 仓库业务代码
## 变更说明 <!-- 简述本次改动做了什么、为什么 --> ## 影响范围 - [ ] Client - [ ] Manager - [ ] Agent / Swarm - [ ] CodeGW - [ ] 计费 - [ ] 密钥 / secret_ref - [ ] 审计 - [ ] 发布链路 - [ ] 文档 / 标准 ## 是否读取标准 - [ ] 已读取当前仓 CLAUDE.md - [ ] 已读取当前仓 PROJECT_STANDARD.md - [ ] 已读取 heicodeDocs 相关标准 ## 是否涉及接口契约 - [ ] 不涉及 - [ ] 涉及，已更新 docs/integration 或相关文档 ## 验收方式 <!-- 请说明如何验证，例如： python scripts/test-runtime-contract.py python scripts/test-merge-smoke.py python scripts/test-workflow-e2e.py --> ## TODO / Mock - [ ] 没有 TODO / mock - [ ] 有，说明原因和后续计划 ## Release 仓库确认 - [ ] 本 PR 不涉及 release 仓库业务代码
xiaohei added 2 commits 2026-06-21 20:00:32 +00:00
fix(agent): 执行 max_tokens 4000→65535 + 明确全仓已本地 clone(修两个执行 bug)
CI / guardrails (push) Successful in 4s
Details
CI / tests (push) Successful in 41s
Details
c5595cf344 
#1 全量重写文件的 files JSON 远超 4000 token → JSON 中途截断("Unterminated string")
   → 任务失败。提到 qwen3.7-max 输出上限 65535,env AGENT_EXEC_MAX_TOKENS 可调。
#2 prompt 明确:完整仓库已 clone 在 {workspace_dir}(base_commit),上面列出的文件只是
   片段,绝不能因没列出就当文件/目录缺失(之前 LLM 误判 qutebrowser/utils 不存在 → 联网)。

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
fix(agent): content 为空时回退 reasoning_content(修思考模型空响应)
CI / guardrails (push) Successful in 6s
Details
CI / tests (push) Successful in 44s
Details
CI / guardrails (pull_request) Successful in 5s
Details
CI / tests (pull_request) Successful in 39s
Details
c4c2116122 
qwen3.7-max 思考模型常返回空 message.content、答案在 reasoning_content。原代码只读
content → 拿到 "" → json.loads("") → "Expecting value: line 1 column 1 (char 0)" → 任务失败。
新增 _message_text():content → reasoning_content(attr 或 OpenAI SDK model_extra)回退,
_complete 两个返回点都用它。

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
xiaohei merged commit b0831645cf into main 2026-06-21 20:00:36 +00:00
Sign in to join this conversation.
Reviewers
No Reviewers
Labels
No items
No labels
Milestone
No items
No Milestone
Projects
Clear projects
No projects
Assignees
Clear assignees
xiaohei
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: xiaohei/Agentswarm#24