qakit

Turn a feature an AI agent just implemented into a manual QA plan a human can actually run, with concrete steps, verification checkpoints inline, and pass/fail/skip boxes, grounded in what the code actually changed rather than a generic checklist. The plan is written to docs/qa/ so a person can walk it top to bottom and sign off on the feature by hand.

This is manual QA, meaning steps that genuinely need a human to perform and judge (click through a flow, read a screen, feel out the UX), not checks a machine can do on its own. The value is a disciplined, diff-grounded plan: what to test, in what order, with what setup, and how to know it passed, focused on the things that can't be verified without a human eye.

Anything an AI agent or script can confirm on its own, such as running a terminal command and reading its output, hitting an endpoint, or asserting a return value, does not belong in the human's checklist. The agent runs those itself and reports the results in an Automated verification section at the end of the plan, so the human sees them confirmed without re-running them by hand.

The plan is organized around setup, not around dimensions. Setup is what a manual QA pass actually costs: reseeding data, logging in, getting the app into a particular state. A plan that makes the tester rebuild the same state five times because five cases each restate it is a worse plan than one with fewer, fatter cases.

When this fires

The user finishes building something and wants to verify it by hand: "write a QA plan", "manual test plan", "how do I test this feature", "give me a testing checklist", "/qakit", or a bare "QA this" after a coding session. qakit produces a plan a human runs, and it never writes or runs test code. If the user wants automated tests (unit, integration, E2E), that's a separate concern for a test-suite skill; say so and don't produce a manual plan for it.

Procedure

1. Scope the feature

Ground the plan in what was actually built, because a generic plan is worthless. Gather, in order of preference: