- Home
- Skills
- Frontend & Web Apps
- React + TypeScript Review
React + TypeScript Review
Paste a component, hook, diff or page and get a focused code review instead of a wall of nitpicks.
Free
Works with the AI Agents you already use
React + TypeScript Review
Example session with this skill installed
Review this component for bugs
const Search = ({ query }) => {
const [data, setData] = useState(null);
useEffect(() => {
fetch('/api?q=' + query)
.then((r) => r.json())
.then(setData);
}, [query]);
return <div>{data?.name}</div>;
};
- Read your context and instructions
- Compiled the react + typescript
Review: Search
Must fix
Stale response can overwrite newer data (the useEffect)
The fetch has no abort, so if query changes quickly an older request can resolve last and win.
useEffect(() => {
const controller = new AbortController();
fetch(`/api?q=${encodeURIComponent(query)}`, { signal: controller.signal })
.then((r) => r.json())
.then(setData)
.catch((e) => { if (e.name !== "AbortError") console.error(e); });
return () => controller.abort();
}, [query]);
Should fix
- Query is not encoded:
'/api?q=' + querybreaks on&,#or spaces. The fix above usesencodeURIComponent. - No HTTP error handling:
fetchdoes not reject on 4xx/5xx; checkr.okbeforer.json(). - Loose state type:
useState(null)infersnull. UseuseState<{ name: string } | null>(null).
Optional
- Add loading and error states so the UI is not blank while waiting.
Connects securely to your tools. The creator never sees your data.
About this skill
Paste a component, hook, diff or page and get a focused code review instead of a wall of nitpicks.
What it checks
- Correctness: effects that should be plain render logic, missing cleanup and abort handling, wrong dependency arrays, state mutation, unstable list keys, stale closures.
- Types:
any, unchecked casts and!assertions, props that allow impossible combinations. - Next.js:
"use client"placed too high, secrets reachable from client code, data fetched in effects when the server could do it. - UX and accessibility: clickable divs, unlabeled controls, missing loading/error/empty states, modal focus handling.
- Performance: only with a stated reason. It does not push memo/useMemo by default.
How it reports
Findings are grouped into Must fix, Should fix and Optional. Each one has the location, why it matters in practice and a minimal code fix. If the code is fine, it says so.
Notes
- Instructions only: no scripts, no network access, no file writes.
- It adapts to your setup (React Compiler, App Router vs Pages Router, TypeScript strictness).
- It reports only what it can point to in the code and labels guesses as questions.
MIT licensed.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Frontend & Web Apps.
Does not perform runtime profiling. Cannot verify behavior for external API dependencies not present in the code context.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 5 days ago
- Free with an Agensi account
Nothing to install. It runs in your AI agent.
Works best when you point it at specific files or a diff. For Next.js reviews, mention whether you use App Router or Pages Router, and whether the React Compiler is enabled.
Permissions
File Scopes
Read-only. The skill only reads the files you point it at (components, hooks, diffs and the types they import). It does not run commands, write files, use the network or read environment variables.