菜单
Use after changing prompts, skills, the compiler, or editorial policies — or when the user asks to measure edit quality, run regression evals, check agreement with approved edits, or says '評価して', '一致率を測って', 'eval を回して'. Compares agent output against human-approved golden projects.
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
| name | evaluate-edit |
| description | Use after changing prompts, skills, the compiler, or editorial policies — or when the user asks to measure edit quality, run regression evals, check agreement with approved edits, or says '評価して', '一致率を測って', 'eval を回して'. Compares agent output against human-approved golden projects. |
| metadata | {"filePattern":["reports/eval/**"],"bashPattern":["scripts/eval.ts","npm run eval"]} |
project_state.yaml に approval_record.approved_by がある承認済みプロジェクト。approved_by: operator が人間ティアで最も信頼できる正解。docs/eval-harness.md を読む。npm run eval -- --all --min-score 80
reports/eval/）を読み、どのメトリクスが落ちたか特定する。mkdir -p projects/eval-rerun-<name>
cp -R projects/<golden>/01_intent projects/<golden>/03_analysis projects/eval-rerun-<name>/
npm run eval -- --candidate projects/eval-rerun-<name> --golden projects/<golden>
GEMINI_API_KEY 必須）:npm run eval -- --candidate projects/eval-rerun-<name> --golden projects/<golden> --judge
人間編集者が編集を確定したら、そのプロジェクトの project_state.yaml に記録する:
approval_record:
approved_by: operator
approved_at: "<ISO8601>"
status: clean
登録ファイルは不要。npm run eval -- --list で検出を確認する。
reports/eval/<mode>-<candidate>-<timestamp>.md（人間向け）reports/eval/<mode>-<candidate>-<timestamp>.json（機械向け）MUST USE when the user asks to crop, reframe, trim, or improve composition of still photos, especially selected event photos, album photos, portraits, speakers, panels, workshops, or group photos. Use after photo selection or correction when the task is composition improvement rather than selecting images.
MUST USE when the user provides a photo ZIP, photo folder, or still-image event material and asks for photo selection, album selection, highlights, digest photos, curation, or quality filtering. Use this instead of analyze-footage when the source is primarily still photos and the requested work is selecting images rather than editing or video planning.
Use when timeline.json exists and the user asks to review the rough cut, critique the edit, or generate review_report.yaml and review_patch.json.
MUST USE when starting a new video project, when the user provides footage and asks to edit/create a video, or when 01_intent/creative_brief.yaml does not yet exist. This skill interviews the user to understand their intent before any editing begins. Never skip this step.
MUST USE when the user is setting up this project for the first time, when dependencies are missing, API keys are not configured, or when 'npm run demo' fails. Triggers on: 'セットアップ', '環境構築', 'setup', 'install', 'API key', 'ffmpeg not found', 'GEMINI_API_KEY', 'cannot find module'.
MUST USE when source footage or a material folder is provided and material understanding or 03_analysis generation is needed. Trigger not only on explicit analysis requests ('分析して', '素材を見て') but also when the user points at footage and says things like 'この動画で', 'この素材で', or any request that requires understanding the provided media before planning or editing.