qa

Runs the test suite, classifies failures, applies targeted fixes, and re-runs until clean.

Updated May 27, 2026
One-click install
npx skills add https://github.com/jokerman89/lintel --skill qa-jokerman89
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: qa
Source: https://github.com/jokerman89/lintel/tree/main/skills/qa
Command: npx skills add https://github.com/jokerman89/lintel --skill qa-jokerman89

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After making code changes, developers need to know whether the test suite passes and to resolve common failures without manually triaging every snapshot drift, lint nit, or missing import. This Skill automates that loop while refusing to touch real product-logic bugs. ## Core Features & Use Cases - Automated test execution and failure classification: Detects the test runner (npm test, pytest, cargo test), runs the suite, and buckets failures into snapshot drift, lint nits, type errors, assertion failures, and flaky timeouts. - Targeted auto-fixes with guardrails: Applies safe fixes for snapshots, lint, and missing imports, re-runs up to a configurable iteration limit, and never edits product logic without explicit confirmation. - Structured stuck-state reporting: Produces a clean-pass summary or a detailed escalation report with likely causes and recommendations such as /investigate. - Use Case: After a refactor, run the skill to auto-update stale snapshots and fix lint issues, then receive a report showing two remaining assertion failures that need real investigation. ## Quick Start Ask the agent to run the qa skill to execute the test suite, auto-fix common failures, and report whether the branch is clean or stuck.

Frequently Asked Questions about qa

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automatically fix failing tests after a code change?▼

Run the qa skill to execute the test suite, classify failures, and auto-fix common issues like snapshot drift, lint nits, and missing imports. It re-runs up to a configurable iteration limit and reports a clean pass or a structured stuck state.

What test runners does the qa skill support?▼

The skill detects the runner from package.json, pyproject.toml, or Cargo.toml and picks the canonical script such as npm test, pytest, or cargo test. If multiple runners exist, it asks the operator which to use.

Can I run the test suite without applying any auto-fixes?▼

Yes, pass the --no-fix flag to report failures without editing anything, matching the read-only behavior of the /qa-only variant. This is appropriate for ship-gate verification where no edits are allowed.

Why does the qa skill refuse to fix some test failures?▼

Assertion failures indicating real product behavior changes are never auto-fixed; they are escalated with a recommendation to use /investigate. The skill also blocks fixes touching frozen-zone paths and reverts any fix that introduces a new failure.

When should I not use the qa skill?▼

Avoid it when tests have never passed on the branch, since a green baseline must be established manually first. Also use /investigate instead for hypothesis-driven debugging of product-logic bugs, and /qa-only for production deploy gates.