メニュー
GitHub Copilot code review using thermo-nuclear correctness and code-quality rubrics. Use for pull request review, branch diffs, and review of added or modified code.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
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。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/asteasolutions/ai-toolkit --skill code-reviewコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Grounded research. Traceability-first research workflow for evidence, reliability, product, policy, technical, sentiment, complaint, review, public-opinion, and source-backed questions. Use when the user asks for grounded research, rigorous research, exact source traceability, text fragment links, citations, verification, fact-checking, adversarial review, counterevidence, confidence labels, source hierarchy, or a claim-to-source evidence map. Also use when another skill needs source-backed claims checked against primary sources, current product/tool information, community sentiment, or contested evidence.
Evidence discipline for factual claims. Receipts before claims, no guessing, unknown is an allowed answer. Use as the baseline standard when citing sources, files, or command output, and when the user asks you to re-check an answer you already gave because they cannot verify it themselves.
use when the user wants a simple worded reply, or expresses that your speech is too complex, jargon-heavy, hard to understand etc. Or when they say "keep it simple"
SKILL.md を表示中
| name | code-review |
| description | GitHub Copilot code review using thermo-nuclear correctness and code-quality rubrics. Use for pull request review, branch diffs, and review of added or modified code. |
When reviewing this change, apply BOTH of the following skills as complete rubrics. Do not skip either pass.
If those skills are not already in context, load them by name from sibling skill directories. If they cannot be loaded, still run both passes using the same intent: a harsh correctness/security audit of the diff, then a harsh maintainability audit of the same diff.
Apply any repository instructions (AGENTS.md, .github/copilot-instructions.md, path-specific *.instructions.md) as constraints on findings and remediations. Those files are the source of repo-specific policy. This skill is not.
Prefer a small number of high-conviction comments over a long list of nits.
Priority order:
For each finding include file references, evidence, and a concrete remediation that fits this repository's instructions. Do not rubber-stamp "it works" if the change makes the codebase messier or leaves an obvious simpler structure unused.
Calibrate severity honestly. Do not inflate nits into blockers. Do not downgrade real defects.
