- Home
- Skills
- Wellapp Ai
- Well
- Pr Review
pr-review_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 pr-review- SKILL.md8.0 KB
Overview
This skill automates pre-commit and pre-PR technical validation using shell checks and ReadLints. It verifies runtime services, runs type and lint checks, scans for common quality issues, and computes a risk score to guide the verdict.
How this skill works
It runs a Phase 0 runtime check to ensure dependencies, Docker containers, dev servers, and migrations are in expected states and auto-resolves safe issues. Phase 1 executes typecheck, lint, and ReadLints on changed files. Phase 2 scans for debug statements, TODOs, and hardcoded values. Phase 3 computes a risk score from change metrics and combines results into a PASS/WARN/BLOCK verdict.
When to use it
- Before each commit (automated pre-commit invocation)
- Before pushing a pull request (pre-PR phase)
- Manually during development for a quick validation
- As a gating step when making database or API changes
- When preparing a production-ready PR to reduce regressions
Best practices
- Run full review before pushing a PR; use quick mode only for iterative commits
- Auto-resolve only safe environment issues; perform migrations and server restarts manually
- Fix ReadLints errors and type/lint failures before proceeding
- Remove debug console statements and document TODOs before merge
- Address high risk scores by splitting large changes or adding tests
Example use cases
- Local developer ensures dev servers and database are healthy before committing
- CI or pre-PR pipeline enforces type and lint checks and blocks on ReadLints errors
- Reviewer receives a standardized PR Review Report summarizing runtime, quality, and risk
- Quick validation to confirm no debug logs or hardcoded endpoints were introduced
- Risk scoring helps decide if a change needs a deeper review or rollback
FAQ
The skill warns and does not auto-run migrations. It asks for user confirmation before continuing and marks the review as WARN if user proceeds.
When does the skill block a commit?
It blocks when runtime manual action is required, typecheck or lint fails, ReadLints reports errors, or the user explicitly aborts during server issues.