菜单
Visible Harness analyze entry. Use after experiment runs to turn results into interpretation, decisions, and next-run or writing guidance.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Internal Harness instruction source for auto-iterate-goal. Route through visible Harness aliases or hook contracts instead of invoking directly.
Internal Harness instruction source for auto-paper. Route through visible Harness aliases or hook contracts instead of invoking directly.
Internal Harness instruction source for evaluate. Route through visible Harness aliases or hook contracts instead of invoking directly.
Internal Harness instruction source for iterate. Route through visible Harness aliases or hook contracts instead of invoking directly.
Internal Harness instruction source for orchestrator. Route through visible Harness aliases or hook contracts instead of invoking directly.
Visible Harness write entry. Use for manuscript writing, citation-supported paper work, final documentation, README hardening, and GitHub Pages preparation.
| name | analyze |
| description | Visible Harness analyze entry. Use after experiment runs to turn results into interpretation, decisions, and next-run or writing guidance. |
Use $analyze as the human-facing alias for result analysis. This is not a
separate Skill Contract.
Read and follow:
../../../.agents/skills/evaluate/SKILL.md../../../.agents/references/workflow-guide.md../../../AGENTS.md../../../CLAUDE.mdAnalyze current run artifacts and iteration logs, then produce decision-ready
insights that guide the next $run, a $grill pivot, or $write.
After updating completed run evidence, refresh
docs/30_evidence/Experiment_Evidence_Index.{json,md} using
tooling/evidence/build_experiment_evidence_index.py so $write can consume
experiment results without treating iteration_log.json as the sole source of
truth. If not run, report NOT_RUN with the reason.