verification-loop

Runs build, typecheck, lint, test, security, and diff verification phases before pull requests.

Updated Mar 22, 2026
One-click install
npx skills add https://github.com/diazMelgarejo/orama-system --skill verification-loop-diazmelgarejo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verification-loop
Source: https://github.com/diazMelgarejo/orama-system/tree/main/.cursor/.agents/skills/verification-loop
Command: npx skills add https://github.com/diazMelgarejo/orama-system --skill verification-loop-diazmelgarejo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Code changes often reach pull requests with undetected build failures, type errors, lint violations, failing tests, or leaked secrets. This Skill enforces a structured multi-phase verification loop so quality gates are checked systematically before code is considered done. ## Core Features & Use Cases - Six-Phase Verification Pipeline: Sequentially runs build verification, type checking (tsc/pyright), linting (npm lint/ruff), test suites with coverage thresholds, security scans for secrets and console.log statements, and git diff review. - Structured Verification Report: Produces a standardized PASS/FAIL report across all phases with an overall READY/NOT READY verdict and a list of issues to fix. - Continuous Mode: Supports periodic re-verification during long sessions, triggering checks after each function, component, or task completion. - Use Case: After refactoring a TypeScript module, invoke the skill to confirm the project builds, types check cleanly, tests pass with at least 80% coverage, and no API keys were accidentally committed before opening a PR. ## Quick Start Ask the AI to run the verification loop on the current project and report whether the changes are ready for a pull request.

Frequently Asked Questions about verification-loop

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

FAQPage Schema
How do I verify code quality before creating a pull request?

Run a verification loop that checks build success, type errors, lint warnings, test results with coverage, and security issues, then reviews the git diff. The skill outputs a report marking the changes as READY or NOT READY for a PR.

How to run type checking and linting together in one workflow?

The verification loop chains npx tsc --noEmit or pyright for type checking with npm run lint or ruff check for linting in sequential phases. Each phase reports errors, and critical failures stop the pipeline before continuing.

Does this verification workflow support Python projects?

Yes, it supports Python through pyright for type checking and ruff check for linting, alongside JavaScript and TypeScript tooling like tsc and npm scripts. The appropriate commands are selected per project type in each phase.

What test coverage threshold does the verification check enforce?

The test phase targets a minimum of 80% coverage and reports total tests, passed, failed, and coverage percentage. Projects falling below the threshold are flagged in the final verification report.

Why does the security scan check for console.log statements?

Leftover console.log statements in TypeScript source files often indicate debug code that should not ship to production. The scan greps src/ for console.log alongside checks for leaked secrets like sk- tokens and api_key strings.

When should I not rely on this verification loop?

It complements but does not replace immediate PostToolUse hooks, which catch issues in real time. It also only covers the six defined phases, so domain-specific checks like performance profiling or accessibility audits require separate tools.