Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    qa-pilot Pro — QA audits for web projects

    1

    qa-pilot Pro contains everything in Free and adds a third command, `audit`, for questions that go beyond a single task.

    $9

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    About this skill

    qa-pilot Pro contains everything in Free and adds a third command, audit, for questions that go beyond a single task. Three commands in total: onboard once, test on every task, audit for the big picture.

    What Pro adds to Free

    Free works on one task at a time. Pro answers the questions a single task can't.

    • A roadmap for the whole project. audit rates how mature your test setup is, takes in the test-strategy check, audits your critical flows, and turns it all into one prioritised plan.
    • Requirements checked against the real code. It finds missing test cases, contradictions and bugs in a feature's spec, not only in a single diff.
    • Proof about the tests you already have. It finds false greens and flaky tests, and on a sample it breaks the code to show which tests can really fail.
    • A component library audit. It finds components and states that nothing tests, bottom-up, and can write the missing stories and tests after your yes.
    • 100 more QA rules. Security (38), API/HTTP (23), AI systems (15), auth (14) and requirements (10), on top of the 80+ in Free.

    How audit works

    audit picks the audit from what you give it:

    • Nothing: the whole project. Rates how mature the test setup is: layers, CI gates, mocking, test data, selectors, flaky handling, linting. You confirm the 3–7 critical user flows, and it audits each one. The result is one roadmap plus an action list an agent can execute.
    • A feature, spec or ticket: the requirements. Checks every claim against the code and existing tests. For every entity with a lifecycle, it fills an entity card of must-check web patterns. You get a bucketed action list: Must, Should (at most 8) and Highlight. Bugs to fix come first, then tests with a level and a target, then product questions. Output is Markdown plus JSON, with a P1/P2/P3 evidence report. Three depths: critical, standard and thorough.
    • "tests": the test suite. Finds false greens, dead tests, flakiness sources, copy-paste and over-complexity. On a sample (critical flows plus suspects), it breaks the code under test with a restoring script to prove whether each test can fail. You get a severity-ranked report and an ordered refactoring plan, which it can then carry out step by step.
    • "ui": the component library. Goes bottom-up, atoms first, then compositions. It finds components with no stories or tests and states that are never exercised.

    What's included

    • Three commands: onboard, test and audit.
    • 180+ QA rules: everything in Free, plus API/HTTP, auth, security (OWASP ASVS / WSTG based), AI-system and requirements rule sets.
    • Five stdlib Python scripts: failure probe, profile drift, UI inventory, spec lint, action-list validator.

    Requirements

    A coding agent with terminal access, Git and Python 3. For browser runs, a Playwright MCP server. Skill texts are in English; the rule texts are in Russian. A project audit costs roughly 250k tokens per flow, and you choose which flows to audit.

    qa-pilot reports bugs and does not change product logic. The one exception: during browser testing, on your request, it fixes small, low-risk frontend bugs (typo, broken handler, style). Auth, data and backend issues are only reported.

    Licence

    Personal commercial licence for the buyer: any use, including work for an employer or clients, on any number of projects; not transferable, no redistribution. A company licence is available by email.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Testing & Debugging.

    Reports bugs; does not change product logic, except small low-risk frontend fixes during browser testing when you ask. Needs a one-time onboard run that writes docs/qa/project-profile.md. Built for web projects. Rule texts are in Russian. A project audit costs roughly 250k tokens per flow.

    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 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

    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 17 days ago

    Software
    Python 3GitPlaywright MCPNode.js@playwright/mcp

    Permissions

    Terminal / Shell
    Read Files
    Browser
    Network Access
    Write Files

    Terminal: runs your project's tests and the pack's stdlib Python scripts. Files: reads the project; writes tests, e2e config and fixtures, test ids in components, and reports under docs/qa/; small frontend fixes only on request during browser testing. Browser: only browser testing, through Playwright MCP. Network: only when you approve installing test tools (e.g. Playwright) during e2e setup, and the browser opens your app's URL.

    Listed18 days ago

    What's inside

    Frequently Asked Questions

    Every new test goes through a failure probe: the code under test is broken, the test must go red, and the code is restored. Tests are written to the conventions in your project profile, and coverage is proven with a test, a distinguishing assertion and the CI layer that runs it.

    Whatever your web project already uses. onboard detects the stack, test frameworks and CI and writes them to the profile; e2e tests are Playwright. qa-pilot does not bring its own framework.

    Run onboard once. It reads the stack, conventions and CI, asks what it cannot detect, and writes docs/qa/project-profile.md. Every other skill reads that file.

    Browser testing drives a real browser through a Playwright MCP server that you connect to your agent; the pack prints the config if none is connected. It walks the flow, records repro steps and reports findings.

    Mostly no. It reports bugs and writes tests that stay red until the bug is fixed. Only when you ask during browser testing does it fix small, local frontend bugs, then re-check them in the browser and run the review loop.

    The whole pack: the three commands, 180+ QA rules, five stdlib Python scripts and a personal commercial licence. Versions you received stay usable without a time limit.

    New in Testing & Debugging