菜单
Code review staged changes or a specific area of the codebase, optionally delegating to a chosen agent. Use when the user wants a code review.
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
Reads the diff between the current branch and main and produces a production pre-flight checklist of everything that must be done or configured once the change merges. Use before merging or deploying a branch.
Suggest tests worth writing for a feature using the ZOMBIES heuristic (Zero, One, Many, Boundaries, Interface, Exceptions, Simple scenarios). Pass a free-text feature description, or omit args to use the current branch's diff. Outputs only the categories that apply — not a full ZOMBIES checklist.
🌡️ Evaluate every dependency a branch pulls in — client or server, any language — against the WARM check (Worth it, Alive, Right-sized, Maintained securely). Diffs the branch against a base, finds newly added or upgraded direct dependencies across all manifests, and scores each one. Use when the user asks to "WARM check" a branch, vet new dependencies, or review what a PR adds to the dependency tree.
Triage a branch or diff by grouping changed files into feature areas, assigning each a risk tier (High/Medium/Low), and producing a scannable summary that helps decide where to spend review time. Use when the user asks to "triage" a branch, PR, or diff.
Scan a branch or diff against the First Five checklist (Error Handling, Input Boundaries, External Calls, State Mutations, Assumed Dependencies) and report only genuine concerns. Use when the user asks to "run the first five" on a branch or diff.
Prepare a structured, scannable review checklist for a branch or diff, grouped by feature change and applying the four-pass review order (Types, Data Flow, Business Logic, Edge Cases) within each feature. Use when the user asks to "prep a review", "review order", or wants a checklist to manually walk through changes.
| name | code-review |
| description | Code review staged changes or a specific area of the codebase, optionally delegating to a chosen agent. Use when the user wants a code review. |
| argument-hint | [agent] [feature] |
| disable-model-invocation | true |
| allowed-tools | Bash(git diff:*), Bash(git log:*), Bash, Read, Grep, Glob, Agent |
Staged diff:
!git diff --cached
Recent commits (for context):
!git log --oneline -10
Raw arguments: $ARGUMENTS
Parse the arguments as follows:
claude or self or empty — perform the review directly in this context (default)codex, aider, goose). The review will be delegated by invoking that command via Bash (see Delegation section).If only one argument is given and it does NOT match a known agent name (claude, self), treat it as the feature instead, with agent defaulting to claude.
git add or specify a feature to review, e.g. /code-review voting functionality." Then stop.Read surrounding source files as needed to understand context. Organize findings by file, then by severity.
Group findings by severity. Each severity level that has findings should be its own heading (e.g. ## 🔴 Critical, ## 🟠 Error, etc.). Within each severity heading, list the findings. For each finding, include:
Omit severity headings that have no findings.
End with a ## Summary section: overall assessment, whether changes look good to merge (or code looks healthy), and a count of findings per severity level.
If the agent is not claude/self/empty, it is treated as a CLI command name. Delegate the review by invoking that command via Bash:
<agent> -q "<prompt>" where <agent> is the CLI command the user specified (e.g. codex, aider, goose)