Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    code review excellence

    1

    Structured code and PR reviews with severity-based findings, security checks, and test quality audits.

    $8

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    code review excellence

    code review excellence

    Example session with this skill installed

    Review this diff for a Node.js API endpoint. It adds a new GET /user/:id route. I've attached the code and the test file. Business objective: fast retrieval for the profile page.

    • Read your context and instructions
    • Compiled the code review excellence

    I have reviewed the changes for the GET /user/:id endpoint. The review focuses on security, performance, and test quality based on the provided diff and test suite.

    Code Review Report: User Profile API

    SeverityCategoryFile:LineFindingRationale
    BLOCKINGSecurityroutes/user.js:14Potential NoSQL InjectionThe id param is passed directly to the find() query without validation.
    IMPORTANTPerformanceservices/userService.js:22Missing ProjectionThe query fetches the entire user object, including password_hash, when only profile data is needed.
    SUGGESTIONMaintainabilitytests/user.test.js:45Hardcoded IDThe test uses a static UUID. Using a factory or dynamic ID improves test isolation.
    PRAISECorrectnessmiddleware/auth.js:5Robust AuthGood use of early returns for unauthorized tokens.

    Detailed Findings

    [BLOCKING] NoSQL Injection Risk

    • Evidence: db.users.find({ _id: req.params.id })

    Rationale: Untrusted input from the URL segment is used in a database query. Attackers could pass query operators to bypass filters.

    • Suggested Change:
    const userId = String(req.params.id);
    if (!isValidObjectId(userId)) return res.status(400).send('Invalid ID');
    db.users.find({ _id: userId });
    

    [IMPORTANT] Data Over-fetching

    • Evidence: The current implementation returns all fields.
    • Rationale: Minimizing payload size improves response times and prevents accidental PII leaks.
    • Suggested Change: Add .select('username email bio avatar') to the Mongoose query.

    Verdict

    REQUEST CHANGES
    The injection vulnerability must be addressed before this is merged.

    Next steps

    1. Sanitize the req.params.id input.
    2. Add a projection to the database query to limit returned fields.
    3. Update the test suite to include a case for invalid ID formats.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    The problem

    Manual code reviews often miss critical security flaws or architectural debt because reviewers get bogged down in syntax nits. Teams struggle with inconsistent feedback quality and lack a structured way to prioritize blocking issues over minor suggestions.

    What it does

    • Analyzes pull requests and diffs to identify logic errors, security risks, and maintenance debt.
    • Categorizes findings by severity: blocking issues, important suggestions, questions, and praise.
    • Evaluates test coverage and quality to ensure changes are verified before approval.
    • Provides actionable recommendations with clear rationales and validation steps.

    Why this beats prompting it yourself

    Generic prompts often result in "LGTM" or shallow formatting critiques. This skill enforces a rigorous severity classification system and prevents the AI from fabricating test results or ignoring missing context, ensuring every finding is backed by evidence from your specific files.

    Use cases

    • Reviewing complex pull requests for architectural alignment and security risks.
    • Auditing test quality and coverage for mission-critical features.
    • Standardizing review comments across a large engineering team.
    • Summarizing technical decisions and risks for stakeholders during a release cycle.

    Known limitations

    Does not execute code or run actual CI/CD pipelines. It cannot replace formal security assurance or automated linting tools for style enforcement.

    Cannot execute code or run live tests. Requires manual input of CI/CD results or test logs for coverage assessment. Does not handle automated formatting.

    How to use

    1. 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. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 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

    Permissions

    No special permissions declared or detected

    Listed2 months ago

    What's inside

    Frequently Asked Questions

    The skill categorizes feedback into specific tiers—blocking issues, important suggestions, questions, and praise—ensuring your team knows exactly what must be fixed before a merge versus what is a minor stylistic preference.

    This skill is designed to work with major AI agents that support code analysis; simply provide the diff or pull request content as input, and the agent will apply the specialized logic to generate the structured review.

    Unlike generic prompts, this skill uses a specialized framework to audit test quality and verify that code changes are actually backed by meaningful tests, rather than just checking for the existence of test files.

    Upon purchase, you receive the full configuration for the skill, including the structured review logic, the severity classification framework, and documentation on how to integrate it into your existing PR process.

    While the skill provides deep logical and security analysis, it does not execute code or replace automated linters; it is best used as a high-level architectural and security gatekeeper alongside your existing CI/CD pipeline.

    New in Code Quality & Review