More screenshots

    Works with the AI tools you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIGitHub CopilotGitHub CopilotGemini CLIGemini CLI+20 more

    Production Code Review Gate

    1

    Most AI code reviews produce long lists of generic warnings.

    $9

    Secure checkout via Stripe

    30-day refund guarantee

    Converts to your local currency at checkout

    0 installsSecurity scanned
    Production Code Review Gate

    Production Code Review Gate

    Example session with this skill installed

    Review this PR diff for a Node.js service adding a new Redis-backed rate limiter. Check for concurrency issues and production readiness.

    • Read your context and instructions
    • Compiled the production code review
    • Generated the document

    I have completed a production readiness review of the Redis rate limiter implementation. The review identified a critical atomicity issue in the increment logic that could lead to limit bypass under high concurrency.

    Production Readiness Report: Rate Limiter PR

    IDFileSeverityConfidenceFindingRelease Impact
    CODE-01src/middleware/rateLimiter.tsHIGH0.95 (Verified)Non-atomic GET/SET sequence allows race condition.BLOCK
    OPS-01docker-compose.ymlMEDIUM0.85 (Partial)Missing Redis resource limits in container config.REVISE
    TEST-01tests/limiter.test.tsLOW0.90 (Verified)Missing test case for Redis connection timeout.REVISE

    Findings & Evidence

    CODE-01: Race Condition in Increment Logic
    The implementation fetches the current count, increments in-memory, and then sets the value back to Redis. Under concurrent requests, multiple workers will read the same initial value, causing the counter to under-count.

    • Evidence: Verified via inspection of lines 42-45. No INCR or Lua script usage found.
    • Remediation: Use the Redis INCR command or a Lua script to ensure atomicity.

    OPS-01: Resource Constraints
    The Redis service definition lacks deploy.resources.limits. In a production environment, an unconstrained Redis instance could exhaust host memory during a spike.

    • Evidence: Observed missing keys in services.redis.

    Gate Decision: BLOCK

    Next steps

    1. Replace the manual GET/SET flow in rateLimiter.ts with an atomic INCR operation.
    2. Add memory limits to the Redis container definition.
    3. Add a concurrency test to verify the rate limit holds under parallel execution.

    production-code-review-gate.pdf

    PDF · document

    Generated

    Example file from a real run - the skill writes it into your workspace.

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

    What you get

    Audit complex PRs for race conditions and distributed state errors.Validate SQL migrations for breaking changes and lock risks.Perform security-focused reviews on trust boundaries and API contracts.Enforce a consistent production-readiness gate across a development team.Review pull requests before merge or production deployment.Detect correctness, security, performance, and concurrency risks.Generate evidence-backed PASS, REVISE, or BLOCK release decisions.Identify missing tests and propose focused regression coverage.Review dependency, configuration, and rollout risks.Produce file-organized findings with severity and confidence scores.

    About this skill

    Most AI code reviews produce long lists of generic warnings. Production Code Review Gate is built for the harder question: can this change safely ship?

    AG-1 reviews repositories, pull requests, diffs, file sets, and pasted code using a repeatable evidence gate. It maps the changed surface, follows reachable callers and failure paths, checks counter-evidence, and publishes a finding only when it can identify the cause, trigger, mechanism, and production impact. Unverified concerns are separated into a Watchlist instead of being presented as bugs.

    Each finding is organized by file and includes CRITICAL, HIGH, MEDIUM, or LOW severity; a calibrated confidence score; code-grounded evidence; a concrete remediation; an optional safe patch proposal; and focused tests that prove the correction.

    The final production-readiness report covers functional correctness, security, reliability, concurrency and data integrity, performance, compatibility and rollout, tests and observability, dependencies and configuration, and maintainability.

    AG-1 supports application code, APIs, databases, queues, infrastructure, containers, CI/CD, and polyglot repositories. Its dedicated SQL protocol covers injection boundaries, NULL logic, join cardinality, pagination, transactions, isolation, deadlocks, query plans, indexes, connection pools, and safe migrations.

    It understands partial context, avoids fake CVEs and invented test results, redacts exposed secrets, and never treats formatting preferences as release defects.

    Every review ends with one decision:

    PASS — no required remediation remains in the reviewed scope. REVISE — changes or additional evidence are required before release. BLOCK — a sufficiently evidenced critical production risk must stop the release.

    Ideal for senior developers, tech leads, consultants, agencies, startup teams, and engineering managers who need a serious pre-merge or pre-release quality gate.

    How to install

    Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.

    ~30 seconds
    1. 1

      Download the ZIP

      Free skills download straight away. Paid skills unlock right after purchase.

    2. 2

      Unzip into your skills folder

      Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.

    3. 3

      Ask your agent to use it

      Restart the agent if it was already running. It picks the skill up automatically - no config needed.

    Skills folder by agent

    Click the path to copy it. Create the folder if it does not exist yet.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill passes our trust checks below.

    Security scanned

    Passed our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    30-day refund

    Not a fit? Get your money back

    Trust & safety

    Security scanned

    Verified clean 26 days ago

    • Passed all security checks, Safe to install

    Listed26 days ago
    Updated15 days ago

    What's inside

    Frequently Asked Questions