Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Test Flake Hunter

    2

    Detect and analyze flaky tests across multiple frameworks with automated repeated execution and severity reporting.

    $7

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    test-flake-hunter

    test-flake-hunter

    Example session with this skill installed

    I've noticed some intermittent failures in our authentication tests. Can you run the tests in tests/auth/ 20 times and tell me which ones are flaky?

    • Read your context and instructions
    • Compiled the test-flake-hunter

    Found 1 flaky test in tests/auth/:

    • test_session_expiry: 14 passes, 6 failures (30% flake ratio).
      Severity: High.
      Suggested cause: Race condition in session cleanup timing.
      Detailed reports saved to flake-report.md and flake-report.json.
    Detailed reports saved to flake-report.md and flake-report.json

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

    About this skill

    What it does

    Test Flake Hunter is a diagnostic tool designed to identify non-deterministic test failures. It automatically detects your testing framework—supporting pytest, Jest, and Go test—and executes your suite or specific test files multiple times to uncover "flaky" behavior. By comparing pass/fail patterns across runs, it calculates flake ratios and provides structured reports on test reliability.

    Why use this skill

    Manually re-running tests to catch intermittent failures is tedious and prone to human error. This skill automates the repetition, normalizes output from different runners, and applies severity scoring to help you prioritize fixes. It goes beyond simple "pass/fail" by analyzing error messages and execution patterns to suggest likely root causes, such as network timeouts or race conditions.

    Supported Tools

    • Python: pytest, py.test
    • JavaScript/TypeScript: Jest
    • Go: go test
    • Generic: Any runner with standard exit codes via Makefile or custom commands

    Output format

    The skill produces two primary artifacts: a human-readable flake-report.md summary for quick review, and a structured flake-report.json for integration into CI/CD pipelines or further data analysis. Reports include flake ratios, failure message snippets, and suggested remediation steps.

    • Expensive for large suites due to N repetitions.
    • Side effects in tests can cause false positives.
    • Filter syntax varies by runner.

    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 4 months ago

    Permissions

    Terminal / Shell
    Read Files
    Listed4 months ago

    What's inside

    Frequently Asked Questions

    This skill automates the detection of non-deterministic tests by repeatedly executing suites across Pytest, Jest, or Go, calculating "flake ratios" to help you distinguish between genuine bugs and intermittent infrastructure or race-condition failures.

    The skill is designed to run within any environment where your test runners are installed; it works with major AI agents that have access to a terminal or CLI, such as AutoGPT, OpenDevin, or custom LangChain agents.

    Yes, while it has native support for Pytest, Jest, and Go, you can use the 'Generic' mode to wrap any command-line test runner (like those in a Makefile) that utilizes standard exit codes.

    Your purchase includes the full skill logic and documentation; once added to your toolkit, you will receive notifications for updates and improvements to the detection algorithms at no extra cost.

    Setup is minimal; once you grant the agent access to your repository, you simply provide the path to your test suite, and the skill handles framework detection and execution parameters automatically.

    The skill generates a human-readable Markdown summary for developers and a structured JSON file containing flake ratios and failure patterns, making it easy to integrate into existing CI/CD dashboards.

    New in Testing & Debugging