piv-validate

Runs project tests, type checks, and linting to report a single pass or fail verdict.

Updated Aug 28, 2026
One-click install
npx skills add https://github.com/az9713/claude-code-hooks-tutorial --skill piv-validate-az9713
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: piv-validate
Source: https://github.com/az9713/claude-code-hooks-tutorial/tree/main/skills/.claude/skills/piv-validate
Command: npx skills add https://github.com/az9713/claude-code-hooks-tutorial --skill piv-validate-az9713

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Before committing or opening a pull request, developers need to confirm that nothing broke across the whole stack, but manually running tests, type checks, and linters for every surface is slow and easy to skip. This Skill runs the full validation suite in one pass and reports a single PASS/FAIL verdict. ## Core Features & Use Cases - Full-stack validation: Runs tests, type checks, and lint/format checks across every surface of the project, continuing after failures so the report covers everything. - Structured summary report: Produces one line per check with a pass or fail marker, plus the failing command and output for every failure, without attempting fixes. - Customizable command template: Ships as a placeholder template you replace with your project's real commands from package.json, pyproject.toml, Makefile, or CI configuration. - Use Case: After finishing a feature touching both backend and frontend, invoke the validator to run backend pytest, mypy, and ruff plus frontend vitest, tsc, and eslint, then get one overall verdict before opening the PR. ## Quick Start Ask the assistant to run the full project validation suite and report whether all tests, type checks, and lint checks pass before committing.

Frequently Asked Questions about piv-validate

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

FAQPage Schema
How do I run all project checks before committing code?▼

Use a validation skill that runs your test suite, type checker, and linter in sequence and reports a single PASS or FAIL verdict. Configure it with the real commands from your package.json, pyproject.toml, or Makefile so one invocation covers the whole repo.

How do I set up a validation skill for a full-stack project?▼

Add one section per surface: backend tests, backend types, backend lint, then the same for frontend. Find the real commands in package.json scripts, pyproject.toml, or your CI workflow, and run each from the directory where its config file lives.

Why does my type check pass locally but fail in CI?▼

Many tools only discover their config from the current working directory. If your config lives in backend/pyproject.toml, running from the repo root silently skips it, so the command must cd into the correct directory first.

Should a validation script stop at the first failure?▼

No, it should continue after failures so the final report covers every check at once. Capture the output of each failing command and include it in the summary so all problems are visible in a single run.

When should I add a live smoke test to validation?▼

Only add a smoke test when changes touch routing, middleware, or startup code, and skip it when the test suite already exercises the app in-process. Prefer starting the server in a second shell over backgrounding and killing it, which is not portable.