- Home
- Skills
- Wellapp Ai
- Well
- Qa Commit
qa-commit_skill
- TypeScript
306
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill wellapp-ai/well --skill qa-commit- SKILL.md4.7 KB
Overview
This skill verifies that the current commit satisfies its assigned QA Contract (G#N Gherkin scenarios and AC#N acceptance criteria) and returns a clear GREEN (pass) or RED (fail) verdict. On RED it automatically invokes the debug skill with contextual failure data so issues can be fixed and verification retried. It is designed for TypeScript FinOps infrastructure repositories where correctness of APIs and UI acceptance matters before committing.
How this skill works
The skill loads the commit context and the QA Contract listing G# and AC# criteria, then performs technical validation: lint/type checks and related test runs. It inspects backend Gherkin scenarios by locating endpoints, response shapes, and error handling, and it verifies frontend acceptance criteria via component/story existence and optional Storybook browser checks. Finally it emits a verification report (PASS/FAIL per criterion) and, on RED, auto-invokes the debug skill with error messages, failing criteria, and any browser console/network evidence.
When to use it
- After PR review passes and before creating the git commit
- During automated agent-mode commit-level workflows
- When manually validating that a change satisfies assigned Gherkin and AC criteria
- Before merging feature branches that affect APIs or UI components
- When you need reproducible evidence (screenshots/logs) to support QA decisions
Best practices
- Ensure the commit plan includes the exact "Satisfies: G#X, AC#Y" list before running the skill
- Run typecheck and unit tests locally first to reduce noisy failures
- Keep Storybook stories and component tests up to date to enable reliable AC checks
- Include clear failure messages and minimal repro steps to speed debug when RED occurs
- Re-run qa-commit after debug fixes until the verdict is GREEN
Example use cases
- Validate an API change implements its Gherkin endpoints and response schema before committing
- Confirm a UI component meets visual/interaction ACs via Storybook snapshots and browser actions
- Gate commits in a CI workflow so only GREEN-verfied changes proceed to merge
- Automatically trigger debug with detailed context when lint/typecheck or tests fail
FAQ
It auto-invokes the debug skill and passes failed criteria, linter/typecheck/test output, and any browser console or network errors. After fixes, re-run the verification.
Can I run only backend or frontend checks?
Yes. You can scope the run by providing the relevant G# or AC# criteria and limiting browser MCP steps if Storybook verification is not required.
What evidence does the verification report include?
The report lists technical check statuses (ReadLints, TypeCheck, Tests), per-criterion PASS/FAIL entries, and links or notes for screenshots, logs, or failing stack traces when available.