integration — report
【免费下载链接】comet-llmDebug, evaluate, and monitor your LLM applications, RAG systems, and agentic workflows with comprehensive tracing, automated evaluations, and production-ready dashboards.项目地址: https://gitcode.com/GitHub_Trending/co/comet-llm
Mode:new | update | maintainLanguage:Python | TypeScriptPattern:method-patching | proxy | callback | OTel exporterEntrypoint:track_<name>(...)Library version prepared:==
What was done
- <创建/修改的文件——按 integration / tests / docs 分组>
- <准备动作：依赖安装、版本固定、fixtures/env>
Verification
- MCP:<project 名 + 读回的 trace ids，或 "not verified — <原因>">
- Tests:<N passing — 列出用例>; ruff/mypy <clean | issues>
Flows supported & test coverage
| Flow | Implemented | Test | MCP-verified |
|---|---|---|---|
chat.complete(sync, non-stream) | ✅ | test_<name>_complete__happyflow | ✅ trace<id> |
chat.complete_async | ✅ | test_<name>_complete_async__happyflow | — |
chat.stream(sync) | ✅ | test_<name>_stream__happyflow | ✅ trace<id> |
chat.stream_async | ✅/❌ | … | … |
nested under@track | ✅/❌ | … | … |
error →error_infologged | ✅ | test_<name>__error... | — |
| token usage captured | ✅ | asserted in above | ✅ |
customproject_name | ✅/❌ | param case | — |
What's NOT supported / limitations
- <有意不打补丁的方法、已实现但未测试的流程、已知缺口、provider 不在 LLMProvider 的注意点、env/后端阻塞>
Follow-ups
- <建议的下一步：更多方法、TS 对应实现、PR 拆分等>
How to use
<最小代码片段>
【免费下载链接】comet-llmDebug, evaluate, and monitor your LLM applications, RAG systems, and agentic workflows with comprehensive tracing, automated evaluations, and production-ready dashboards.项目地址: https://gitcode.com/GitHub_Trending/co/comet-llm
创作声明：本文部分内容由AI辅助生成（AIGC），仅供参考
