ワンクリックで
review
Structured code and infrastructure PR review with evidence-based findings, auto-fix classification, and scope drift detection.
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。
メニュー
Structured code and infrastructure PR review with evidence-based findings, auto-fix classification, and scope drift detection.
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。
End-to-end pipeline to augment any repository for AI-assisted development. Installs opinionated engineering workflows, generates documentation, and configures Claude Code or Cursor.
Destructive command safety guardrails. Warns before dangerous operations in production and shared environments.
Transform a knowledge graph into a human-readable /docs folder with markdown documentation that both humans and agentic workflows can reference.
Systematic root-cause debugging and incident investigation. No fixes without understanding the problem first.
Architecture review before implementation. Walk through design, failure modes, scope, and test strategy before writing code.
Engineering retrospective and velocity analytics from git history. Generates weekly insights on team productivity, code quality, and contribution patterns.
SOC 職業分類に基づく
| name | review |
| version | 1.0.0 |
| description | Structured code and infrastructure PR review with evidence-based findings, auto-fix classification, and scope drift detection. |
| author | iscmga |
| tags | ["code-review","pr-review","infrastructure","terraform","security","quality"] |
| triggers | {"globs":["**/*.tf","**/*.ts","**/*.py","**/*.go","**/*.yaml"],"keywords":["review","code review","pr review","pull request","review changes","review PR"]} |
Structured pre-merge review workflow that catches issues tests miss. Focuses on evidence-based findings — every claim needs a line number, test reference, or concrete proof. Inspired by gstack's review methodology.
ship skill pipeline# Detect base branch
BASE=$(gh pr view --json baseRefName -q '.baseRefName' 2>/dev/null || git symbolic-ref refs/remotes/origin/HEAD 2>/dev/null | sed 's@^refs/remotes/origin/@@' || echo "main")
# Get the diff
git diff "$BASE"...HEAD --stat
git diff "$BASE"...HEAD
If there are no changes, stop. Nothing to review.
Read the PR description or most recent commit messages. Then compare to the actual diff:
If scope drift is detected, report it:
SCOPE DRIFT: The following changes appear unrelated to the stated purpose:
- path/to/file.tf — adds unrelated tagging changes
- path/to/other.py — reformats code not related to the feature
Check for issues that MUST be fixed before merge. For infrastructure repos, focus on:
Security
Correctness
Data Safety
Blast Radius
Lower-priority findings that improve quality:
For each finding, classify as:
| Classification | Action | When |
|---|---|---|
| AUTO-FIX | Apply the fix directly | Obvious, low-risk fixes (typos, missing tags, formatting) |
| ASK | Present to user for decision | Ambiguous, architectural, or potentially breaking changes |
| INFO | Note for awareness | Style, conventions, nice-to-haves |
For each AUTO-FIX item:
Never auto-fix:
## Review Summary
**Branch:** feature/xyz → main
**Files changed:** N
**Scope:** [CLEAN | DRIFT DETECTED]
### Critical (must fix)
| # | File:Line | Issue | Classification |
|---|-----------|-------|----------------|
| 1 | main.tf:42 | IAM policy uses wildcard action | ASK |
| 2 | variables.tf:15 | Missing type constraint | AUTO-FIX (applied) |
### Informational
| # | File:Line | Issue |
|---|-----------|-------|
| 1 | modules/vpc/main.tf:88 | Consider adding description to output |
### Auto-fixes Applied
- variables.tf:15 — added type = string constraint
- tags.tf:3 — added missing Environment tag
### Verdict: [APPROVE | REQUEST CHANGES | NEEDS DISCUSSION]
terraform validate passesterraform fmt -check passes-target usage in CI/CDship skillReview is a required step in the ship pipeline. Ship invokes review before creating a PR.
careful skillReview should flag the same destructive patterns that careful guards against.
security-review skillFor security-focused changes, invoke the full security-review skill instead of the security subset here.