メニュー
Review a commit or PR for issues using multi-agent analysis with confidence-based findings
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
Verify review or audit findings against the false positive guide to eliminate false positives and assign confidence levels
Internal resource bundle for rev-review, rev-audit, and rev-verify skills. Holds shared review principles, language pack detection, agent prompts, and audit focuses. Do not invoke directly — the rev-* skills read its resources via `../load-context/resources/` paths.
Run a full project audit on any codebase with situational focus activation based on user instructions
| name | rev-review |
| description | Review a commit or PR for issues using multi-agent analysis with confidence-based findings |
| disable-model-invocation | true |
! ls package.json tsconfig.json jsconfig.json go.mod Cargo.toml pyproject.toml requirements.txt pom.xml build.gradle build.gradle.kts settings.gradle settings.gradle.kts gradlew gradlew.bat mvnw mvnw.cmd bun.lockb package-lock.json yarn.lock pnpm-lock.yaml composer.json composer.lock artisan symfony.lock phpstan.neon phpstan.neon.dist psalm.xml psalm.xml.dist phpunit.xml phpunit.xml.dist 2>/dev/null
Read the prompt ../load-context/resources/core/review-core.md
Do a deep analysis of the current branch (or provided branch, commit or PR).
If a commit SHA is provided, analyze that specific commit.
If a PR number is provided, use gh pr diff to get the changes.
If no argument is provided, analyze the current branch.
Use ctx7 CLI to validate framework API usage against current documentation.
The review prompt directory is: ../load-context/resources/
