
Verification Loop
Runs build, type, lint, test, security and diff checks on a Claude Code session before a PR
- Repository
- affaan-m/ECC
- Stars
- 295,127
- Licence
- MIT
- Files
- 1 file
What it does
When installed, the agent walks through a fixed checklist of six phases on your code project: it runs the build, type checker, linter and test suite with coverage, scans for leaked secrets and stray console.log lines, and reviews the git diff of changed files. It then produces a short PASS/FAIL report and a verdict on whether the work is ready for a pull request. The instructions are in Spanish, and the commands shown are for Node/TypeScript and Python projects.
Great for
- Developers finishing a feature who want a consistent pre-PR check run by the agent
- Teams using Claude Code who want the same quality gates applied each session
- Developers working in Spanish who want the checklist in that language
Skip it if
- Non-developers, since it assumes a code project with npm/pnpm, git and build tools installed; it does nothing useful outside a codebase
- Projects using other languages or tooling, because the commands cover only JavaScript/TypeScript and Python, and the agent must adapt them itself
- Anyone relying on it as a real security audit: the secret check is a simple grep for 'sk-' and 'api_key' and can miss or wrongly flag secrets
Try asking
- “Run the verification loop on my changes before I open a PR”
- “Check whether this branch is ready to merge: build, types, lint and tests”
- “Do a verification pass after refactoring the payments module”
- “/verify”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
verification-loop.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip verification-loop.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip verification-loop.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Review the commands the agent runs before it executes them, since they run directly in your project folder and the skill does not restrict them
- Adjust the commands to your project's real scripts; the skill assumes npm run build, lint and test exist
- Treat the secret scan as a first pass only and use a dedicated secret scanner for anything sensitive
What's inside
- SKILL.md
Redistributed unmodified from affaan-m/ECC under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.