メニュー
Review git changes with project-specific risk checks layered on top of general senior code review. Use when asked for a code review, merge readiness assessment, or to flag likely regressions.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Plan and execute code changes using a spec-driven workflow with structured documents. Use when making non-trivial changes (multi-file, multi-step, or behavior-impacting) that need structured planning before implementation. Not for trivial single-file edits.
Create and manage changesets for versioning with @changesets/cli. Use when making changes that need a changelog entry, preparing a release, or managing package versions.
Diagnose and fix failed benchmark runs. Use when a benchmark run fails, scores are unexpected, judges show errors, or traces are incomplete.
Build the pnpm monorepo correctly respecting inter-package dependencies. Use when building, verifying changes, or troubleshooting build failures.
Write and run tests using Node.js built-in test runner (node:test) with .mjs files. Use when creating new tests, debugging test failures, or understanding test infrastructure.
Understand the vanilla JS web-report app structure, patterns, and contribution guidelines. Use when modifying the UI, adding new views, or fixing frontend bugs.
| name | code-review |
| description | Review git changes with project-specific risk checks layered on top of general senior code review. Use when asked for a code review, merge readiness assessment, or to flag likely regressions. |
Perform a project-specific review of current changes.
Scope the diff: git status -sb, git diff --stat, git diff.
Classify touched areas and apply risk checks:
| Area | What to Check |
|---|---|
packages/adapters/ | preflight/execute contract, capability metadata, registry entry, no duplicate IDs, no secret leakage |
packages/judges/ | all judges return critical with a boolean default, regex safety, new type in TaskJudge union |
packages/report/ | weight presets sum to 1.0, getDefaultWeights is the single source of truth, no duplicate definitions |
packages/runner/ | imports weights from report package (not local copy), no duplicate judge execution, scoreWeights written to result |
packages/cli/ | --score-mode valid modes match scoring.ts exactly, new params passed through to runner |
packages/taskpacks/ | new enum fields validated, new judge types parsed, supportedTypes array updated |
packages/core/ | new interfaces exported, union types updated, backward compatibility (new fields optional) |
apps/web-report/ | weights match backend, default mode = "practical", i18n in both languages, design tokens used |
tests/ | new features have tests, no tests silently skipped |
Run automated validation: pnpm build && pnpm test.
Output findings ordered by severity, with file/line references.
