| name | exp-results-ingest-local |
| description | Ingest pulled remote artifacts under research_workspace/experiments/<ExpID>/remote_artifacts, summarize latest eval (summary.json), and update research_workspace/00-实验记录.md (index row + detailed section). Use when you want to record <主指标>/<辅助指标> into 00-实验记录.md after pulling remote artifacts. |
| metadata | {"short-description":"Local ingest + experiment record update"} |
Experiment Results Ingest (Local)
Lightweight Claude mirror of the Codex skill. Full implementation and current script live under:
.codex/skills/exp-results-ingest-local/.
Evidence Rules
- Read artifacts from
research_workspace/experiments/<ExpID>/remote_artifacts/.
- Update
research_workspace/00-实验记录.md.
- Repeated eval summaries are not always repeated seeds. If eval dirs encode
iter* or checkpoint*, record them as checkpoint rows and do not summarize them as mean±std or seed evidence.
- For checkpoint eval sets, “主结果” is the highest checkpoint, for example checkpoint15000, with all checkpoints listed separately.
- For true repeated seed runs,
mean±std is allowed and must be labeled as seed-level evidence.
Usage
python3 .codex/skills/exp-results-ingest-local/scripts/ingest.py --expid "E2025xxxx-xx" --dry-run
python3 .codex/skills/exp-results-ingest-local/scripts/ingest.py --expid "E2025xxxx-xx" --backup
python3 .codex/skills/exp-results-ingest-local/scripts/ingest.py --expid "E2025xxxx-xx" --baseline-expid "E2025yyyy-yy"
python3 .codex/skills/exp-results-ingest-local/scripts/ingest.py --expid "E2025xxxx-xx"
Verification
- Start with
--dry-run and inspect the diff before writing.
- Confirm checkpoint eval sets render as checkpoint rows, not
<主指标>(mean±std).
- Confirm true seed repeats still render as seed-level mean/std.
References
docs/SKILLs/agent-skills-standard.md.codex/skills/exp-results-ingest-local/SKILL.md (full documentation)
