verify

Runs tests, linters, formatters, and type checkers to produce a read-only code verification report.

2|Updated Apr 4, 2026
One-click install
npx skills add https://github.com/DoctorMozg/claude-pipelines --skill verify-doctormozg
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verify
Source: https://github.com/DoctorMozg/claude-pipelines/tree/main/plugins/mz-dev-pipe/skills/verify
Command: npx skills add https://github.com/DoctorMozg/claude-pipelines --skill verify-doctormozg

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Before shipping or reviewing code, you need a trustworthy snapshot of whether tests pass, lint is clean, types check out, and coverage is adequate — without anything being silently auto-fixed. This Skill orchestrates that full verification pass and reports findings without modifying your code. ## Core Features & Use Cases - Multi-tool execution: Detects and runs your project's test framework, linter, formatter, and type checker in parallel waves, including runnable examples and README code blocks. - Coverage and quality analysis: Dispatches reviewer agents to find untested code paths, weak assertions, and fragile tests within the chosen scope. - Failure diagnosis: Groups test and type failures by root cause (logic bug, test bug, config, environment) so you understand why things fail, not just that they fail. - Use Case: Before opening a pull request, run a scoped verification on your branch to get a pass/fail verdict with a written report, then hand failures to a fix-oriented workflow. ## Quick Start Ask the assistant to verify the current branch's code quality and test coverage, for example by saying "verify scope:branch".

Frequently Asked Questions about verify

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

FAQPage Schema
How do I run tests, lint, and type checks in one pass?

Invoke the verify skill with an optional scope such as a branch, path, or free-text focus. It detects your project's test framework, linter, formatter, and type checker, runs them in parallel waves, and compiles a single pass/fail report.

How do I check test coverage and test quality for specific files?

Pass a path or glob like "src/auth/" as the argument. The skill resolves matching source and test files, then dispatches a reviewer agent that reports coverage gaps, weak assertions, and fragile tests limited to that scope.

Does the verification pipeline fix failing tests automatically?

No. The pipeline is strictly read-only and never modifies code. It diagnoses root causes of failures and reports them; fixing is delegated to separate fix-oriented workflows like polish or debug.

What happens if no test framework or type checker is detected?

If no test command is detected, the skill asks you how to run tests rather than skipping. Missing linters, formatters, or type checkers are noted in the report and those checks are skipped without suggesting new tooling.

Why does the verification report say FAIL when only lint warnings exist?

It should not. The verdict logic is binary: test failures or reproducible example failures force FAIL, while lint, format, type, or coverage issues on an otherwise green run remain PASS with Nit-severity findings.