메뉴
Check LangSmith traces for YAMLGraph runs. Use when: debugging graph execution, reviewing LLM latency or token usage, investigating failed runs, verifying tracing config, querying recent traces, troubleshooting LangSmith connectivity.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | check-langsmith-trace |
| description | Check LangSmith traces for YAMLGraph runs. Use when: debugging graph execution, reviewing LLM latency or token usage, investigating failed runs, verifying tracing config, querying recent traces, troubleshooting LangSmith connectivity. |
| argument-hint | optional run ID, 'failed', 'status', or graph path to run first |
Inspect LangSmith traces for YAMLGraph graph runs — config verification, recent run listing, detailed trace analysis, and failure investigation.
Run the status check to confirm tracing is active:
cd /Users/sami.j.p.heikkinen/src/yamlgraph
source .venv/bin/activate
python scripts/langsmith_traces.py status
Expected: Tracing active: True, project name shown, latest run timestamp.
If tracing is not active, check .env for these variables:
LANGSMITH_TRACING=trueLANGSMITH_API_KEY=lsv2_pt_...LANGSMITH_ENDPOINT=https://eu.api.smith.langchain.comLANGSMITH_PROJECT=pr-showcaseBased on the user's argument:
| Argument | Action |
|---|---|
| (none) | List latest 5 runs |
status | Config check only (step 1) |
failed | List recent failed runs |
| a UUID | Show detail + children for that run |
a .yaml path | Run the graph first, then show its trace |
python scripts/langsmith_traces.py list --limit 5
For failures only:
python scripts/langsmith_traces.py list --failed
Get full detail (inputs, outputs, tokens, latency):
python scripts/langsmith_traces.py detail <run-id-or-latest>
Show child node executions:
python scripts/langsmith_traces.py children <run-id-or-latest>
If the user provides a graph path, execute it and then inspect:
yamlgraph graph run <graph.yaml> --var key=value --full
The trace URL prints automatically (🔗 Trace: ...). Then query that run:
python scripts/langsmith_traces.py detail latest
python scripts/langsmith_traces.py children latest
For ad-hoc queries not covered by the script, use the SDK directly:
from dotenv import load_dotenv; load_dotenv()
from langsmith import Client
import json
client = Client()
# Latest run with full outputs
runs = list(client.list_runs(project_name="pr-showcase", is_root=True, limit=1))
run = runs[0]
print(f"Status: {run.status}, Latency: {run.latency}s, Tokens: {run.total_tokens}")
print(json.dumps(run.inputs, indent=2, default=str)[:2000])
print(json.dumps(run.outputs, indent=2, default=str)[:2000])
# Child runs with error details
children = list(client.list_runs(
project_name="pr-showcase",
filter=f'eq(parent_run_id, "{run.id}")',
))
for c in children:
print(f" {c.name} [{c.run_type}] {c.latency}s {c.total_tokens}tok {c.status}")
if c.error:
print(f" ERROR: {c.error[:200]}")
Summarize findings to the user:
eu.api.smith.langchain.com)pr-showcase--share-trace creates a public link (no auth needed)None on error, never raiseAuthor YAMLGraph prompt YAML files. Use when: writing prompt templates, configuring inline schemas or output_schema, using Jinja2 templates, adding system/user messages, defining structured output fields, or using system_segments for prompt caching.
Author YAMLGraph graph YAML files. Use when: creating or editing graph YAML, adding nodes or edges, configuring routing and conditions, setting up state keys, variables, data_files, tools, loop limits, error handling, or verification gates.
Operate the Chaplain automation system. Use when: starting the chaplain, checking pipeline status, submitting proposals to inbox, debugging pipeline failures, retrying failed topics, cleaning worktrees, running the philosopher or inquisitor.
Create and manage YAMLGraph feature requests. Use when: writing a feature request, planning a feature, following the plan-judge-enforce workflow, submitting a proposal to the chaplain inbox, understanding the FR lifecycle, or writing acceptance criteria.
Release a new YAMLGraph version. Use when: bumping version, creating a release, tagging, pushing a release, freezing changelog, or handling pre-commit hook cascades during release.
Run code quality analysis on YAMLGraph. Use when: running ruff, bandit, radon, vulture, pytest coverage, checking code quality, finding dead code, security scanning, or generating improvement recommendations. For autonomous analysis, use the code-analysis agent instead.