acm-observability
Agent-actionable reference for ACM-AI's 6-tool observability stack. Teaches how to query traces, inspect graph state, debug Pydantic failures, and analyze costs programmatically.
npx skills add majiayu000/claude-skill-registry --skill acm-observability --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# ACM-AI Observability Skill ## Decision Tree — Which Tool for Which Problem? | Problem | Tool | Action | |---------|------|--------| | Wrong extraction data | Langfuse | Query trace by `session_id=extraction-{source_id}`, examine LLM input/output spans | | Prompt iteration | LangSmith | Open Playground, edit prompt, re-run side-by-side | | Pipeline costing | Langfuse | Aggregate GENERATION observations by model, sum tokens/cost | | Pydantic parse failure | Logfire (via Langfuse) | Search OTel spans for `pydantic.validate_*` with error status | | Graph stuck / wrong state | LangGraph API | `GET /threads/{id}/state` at `:2024` | | Model relationships | erdantic | Run `scripts/generate_model_diagrams.py` -> `docs/diagrams/*.svg` | | Nested JSON exploration | JSON Crack | Paste JSON at `localhost:8888` | | Pipeline healthy across runs? | Langfuse | Historical traces, score trends, session list | ## Langfuse Query Patterns ### Authentication ```bash # All Langfuse API calls use HTTP Basic auth curl -u "$LANGFUSE_PUBLIC_KEY:$LANGFUSE_SECRET_KEY" \ "$LANGFUSE_BASE_URL/api/public/traces?sessionId=extraction-{source_id}" ``` Default `LANGFUSE_BASE_URL` is `http://localhost:3000` (self-host
- Decision Tree — Which Tool for Which Problem?
- Langfuse Query Patterns
- Authentication
- Python SDK
- Common Filters
- Session ID Convention
- LangGraph API Patterns
- Start the Server
- Common Endpoints
- Swagger UI
- Logfire Safety Guardrails
- NEVER Call instrumentpydantic() Without include={}
- Safe Instrumentation Set
- OTel Span Nesting
All Langfuse API calls use HTTP Basic auth
curl -u "$LANGFUSE_PUBLIC_KEY:$LANGFUSE_SECRET_KEY" \
uv run langgraph dev --no-browser
List registered graphs
curl -s http://127.0.0.1:2024/assistants | python -m json.tool
List threads
curl -s "http://127.0.0.1:2024/threads?limit=10" | python -m json.tool
Get thread state
curl -s http://127.0.0.1:2024/threads/{thread_id}/state | python -m json.tool
Get thread history (checkpoints)What does the acm-observability skill do?
Agent-actionable reference for ACM-AI's 6-tool observability stack. Teaches how to query traces, inspect graph state, debug Pydantic failures, and analyze costs programmatically.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill acm-observability --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.