qa-execution
qa-execution is a Usability testing skill for AI agents, published by compozy in the compozy repository on GitHub. Skillspace rates qa-execution D (66/100) and it passed the safety scan; a judge model has also reviewed it.
Runs real-user dogfooding sessions through the product's public interfaces: a persona walks a journey in the browser, takes a thematic tour, probes edges, hunts paper cuts, and reports what a real user would experience. Reads its plan from the living QA docs tree (<qa-docs-path>, default docs/qa/). Use when validating a release candidate, branch diff, migration, or user-facing change against production-like behavior. For planning that tree — personas, journeys, charters, the bug registry — use qa-report; for CI gate runs, AI implementation audits, or integration/security/performance suites, use agent-output-audit.
- Grade
- D
- Quality
- 66 / 100
- Safety
- Passed
- GitHub stars
- 2,371
Placed #2 of 6 in the judged category final for Usability testing.
Our two independent classifiers differ on this skill's category, so treat the placement as provisional.
This rating describes the version of the repository we crawled; it may have changed since, so go by its current files.
Our review
- Architecture
- 84 / 100
- Maturity
- Structured: ships references or scripts
- Security scan
- No issues found
- License
- MIT
- Adoption
- Adopted by 0 other repositories, in 0 collections
- Scope
- Portable: works in any project. General persona-driven dogfooding QA method with configurable docs path
Judge review
Placed #2 of 6 in the judged category final for Usability testing.
- Trigger
85 85 - Instructions
95 95 - Specificity
88 88 - Safety
92 92 - Disclosure
92 92 - Maintenance
88 88
Anti-fabrication by design: defines a Pass as independently verified and surviving a refresh, marks untestable legs Blocked rather than faking them, and gates every step with an explicit Done-when.
Not checked yet
- No blind task eval yet
- Sibling files (scripts, references) not fetched; security scan covers SKILL.md only
Skillspace indexes and rates this skill; it does not host or endorse it. Run third-party skills at your own risk.