- Home
- Skills
- Code Quality & Review
- Code Review
Works with the AI Agents you already use
Code Review
A strict security and quality gate that scans diffs for secrets, injections, and regressions with zero "vibe" tolerance.
$5
Code Review
Example session with this skill installed
Review the staged changes in src/auth.ts. The test command is 'npm test' and the linter is 'eslint src/auth.ts'. Goal is to ensure no credentials leaked in the new login flow.
- Read your context and instructions
- Compiled the code review
I have completed the security and quality gate for the auth module. The review identified one critical security finding and two non-blocking suggestions.
Code Review Run: Auth Service
| Metric | Value |
|---|---|
| Product | Auth Service |
| Status | FAILED |
| Diff Command | git diff --cached |
| Security Scan | 1 hit (Hardcoded Secret) |
| Quality (Tests) | 0 new failures (Baseline: 0, Current: 0) |
| Independent Reviewer | FAILED |
Verification Findings
- Security [src/auth.ts:42]: Hardcoded assignment to
api_keyusing a string literal. - Logic [src/auth.ts:12]: Missing error handling for the identity provider network request.
- Suggestion: Consider moving the session timeout value to a configuration file.
Reviewer Verdict
{
"passed": false,
"security_concerns": ["src/auth.ts:42 — Hardcoded api_key literal detected"],
"logic_errors": ["src/auth.ts:12 — Potential unhandled promise rejection in login sequence"],
"suggestions": ["Extract timeout constants"],
"summary": "Review failed due to hardcoded credentials and missing error boundaries."
}
Next steps
- Remove the hardcoded API key from
src/auth.tsand use an environment variable. - Wrap the authentication network call in a try/catch block.
- Re-run the review once the fixes are staged.
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
Vague "looks good to me" reviews miss critical security vulnerabilities and regressions. Developers often lack a structured way to enforce hard gates on secrets, injection risks, and new test failures before a commit lands.
What it does
- Scans added lines specifically for secrets, shell injections, unsafe evals, and SQLi.
- Calculates a quality baseline to distinguish between pre-existing failures and new regressions.
- Orchestrates an independent reviewer session to eliminate implementation bias.
- Generates a structured findings report with exact file paths, line numbers, and evidence.
- Enforces a strict "no vibe" policy by failing reviews that lack concrete technical evidence.
Frameworks & tools
Works with any stack providing a CLI interface. Integrates with git, npm test, pytest, ruff, or any buyer-specified test and lint runners.
Why this beats prompting it yourself
Standard prompts often result in chatty, non-deterministic feedback or attempts to fix code mid-review. This skill follows a rigid execution order that prevents "vibe-based" approvals and ensures a fresh context for every review to catch errors the implementer might overlook.
Use cases
- Pre-commit security gating for sensitive modules.
- Automated regression checking against a baseline state.
- Formalizing code quality standards in local development workflows.
- Identifying hardcoded credentials before they reach the remote repository.
Known limitations
Requires the user to provide explicit test and diff commands. Does not support glob patterns for source paths or automated application of suggested fixes.
Requires explicit diff, source, and test commands. No support for glob paths. Only scans added lines in the current diff. Does not apply fixes automatically.
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 any payment.
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
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 2 months ago