kit/plugins/code-review/skills/code-review-agent/SKILL.md
Reviews code for bugs, security issues, and breaking changes. Produces prioritized findings on quality, vulnerabilities, and regressions. Use when asked to review code or check a PR diff.
npx skillsauth add shawn-sandy/agentics code-review-agentInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
When reviewing code, systematically check for common issues across multiple dimensions. Provide specific, actionable feedback with line numbers and code examples. Adapt checklist depth to the code's complexity and context — this is a flexible guide, not a rigid process.
Before reviewing, identify which files to check using this priority order:
Explicit path in message — If the user named a file or directory, use it directly. Skip to the Review Checklist.
Local changes (git status) — If no file was specified, run:
git status --short
Branch diff — Run each in order until files are returned:
git diff main...HEAD --name-onlygit diff master...HEAD --name-onlygit diff HEAD~1 --name-only If files are returned, show the list and
confirm before reviewing. Skip non-reviewable files as above. If all return
empty or fail (e.g., detached HEAD), continue to step 4.Fallback — Ask: "Which file or files would you like me to review?"
Once target files are confirmed, proceed to the Review Checklist for each file.
Read references/review-checklist.md for the full six-dimension checklist. Apply each dimension to every file under review.
Structure the review as follows:
Brief overview of the code's purpose and overall quality (1-2 sentences).
[Low / Medium / High / Very High] — One-sentence rationale (e.g., "Deep nesting in 3 core functions and tightly coupled imports drive the rating.").
List any changes that break existing callers, alter contracts, or risk reintroducing previously fixed behavior. For each:
If none detected: No breaking changes or regression risks identified.
If a breaking change also qualifies as a Critical Issue, list it here only — omit it from Critical Issues to avoid duplication.
Issues that could cause bugs, security vulnerabilities, or data loss. Must be fixed.
Non-critical issues that would improve code quality, maintainability, or performance.
Things the code does well. Reinforce good practices.
See references/example-review.md for a complete sample review demonstrating the expected output format.
development
Turns a React component into a social card with preview, code, and props table. Builds a static preview and screenshots react-card.html via Playwright. Use when asked to share a React component.
data-ai
Refine-prompt: interviews users and assembles a structured AI prompt using Anthropic best-practice techniques. Use when the user runs /plan-agent:refine-prompt or asks to refine a prompt.
development
Plan review Agent Team. Reviews HTML implementation plans in parallel, synthesizes findings, and applies improvements in place. Use when the user asks to review or improve an implementation plan.
data-ai
Craft-prompt: interviews users and assembles a structured AI prompt using Anthropic best-practice techniques. Use when the user runs /plan-agent:craft-prompt or asks to craft a prompt.