verification-runner

Run typecheck, lint, tests, and build in deterministic order.

Updated Jan 16, 2026
One-click install
npx skills add https://github.com/adiomas/claude-code-adiomas-plugins --skill verification-runner
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verification-runner
Source: https://github.com/adiomas/claude-code-adiomas-plugins/tree/main/autonomous-dev/skills/verification-runner
Command: npx skills add https://github.com/adiomas/claude-code-adiomas-plugins --skill verification-runner

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill orchestrates a project verification pipeline by executing typecheck, lint, tests, and build in a deterministic, optimal order to provide fast feedback and high confidence before merges.

Core Features & Use Cases

  • Orderly verification: automatically runs checks in the recommended sequence to catch errors early.
  • Config-driven scope: reads .claude/project-profile.yaml to determine required and optional verifications and whether to stop on failures.
  • Flexible execution: supports full runs and selective verification modes (required, test, lint, build) for fast feedback during development.

Quick Start

Run the full verification: bash scripts/run-verification.sh all

Frequently Asked Questions about verification-runner

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

FAQPage Schema
How do I automate running typecheck, lint, tests, and build in a specific order?

To automate verification in a deterministic order, you can use a pipeline that reads configuration from `.claude/project-profile.yaml` and executes checks sequentially via `scripts/run-verification.sh` to catch errors early before merges.

What is the best way to run only lint or tests during development for fast feedback?

For fast feedback during development, you can run selective verification modes like `required`, `test`, `lint`, or `build` instead of a full run, allowing you to target specific checks and control failure handling based on your project profile configuration.

Do I need a project profile configuration file to run the verification pipeline?

Yes, the verification pipeline requires a valid `.claude/project-profile.yaml` file to determine required and optional checks, define execution scope, and set whether the pipeline should stop on failures before executing `scripts/run-verification.sh`.

How does a config-driven verification pipeline handle build failures?

A config-driven pipeline uses the `.claude/project-profile.yaml` file to define failure handling rules, determining whether the verification process stops immediately upon encountering a typecheck, lint, test, or build error or continues running other checks.

What happens if the verification script is missing from the scripts directory?

The verification pipeline requires the presence of `scripts/run-verification.sh` to execute. If the script is missing, the pipeline cannot run the typecheck, lint, tests, and build checks defined in your project profile.