Skip to content
Get packs

Verification Loop

Runs build, type, lint, test, security and diff checks before a Claude Code session claims work is done

Source Download skill
Repository
affaan-m/ECC
Stars
295,127
Licence
MIT
Files
2 files

What it does

When installed, the agent works through a fixed checklist before saying a coding task is finished: it builds the project, runs a type checker, a linter, and the test suite (looking for an 80% coverage target), scans the code for leftover secrets and console.log calls, and reviews the git diff. It then writes a short PASS/FAIL report with a READY or NOT READY verdict for a pull request. The skill is a set of written instructions; it does not add new tools of its own.

Great for

  • Developers using Claude Code who want a consistent pre-PR quality check
  • Teams that want the agent to report test, type and lint status in a standard format
  • Projects that already have build, lint and test commands set up in npm, pnpm, TypeScript or Python tooling

Skip it if

  • Non-programmers or projects with no build or test setup, since most steps will fail or be skipped
  • Security assurance: the secret check is a simple grep for patterns like 'sk-' and 'api_key' and will miss many leaks or flag harmless text
  • Projects using languages or tools other than the JavaScript/TypeScript and Python examples given, unless the agent adapts the commands itself

Try asking

  • “Verify your work before I open a pull request for this feature”
  • “Run the full verification loop and give me a PASS/FAIL report”
  • “Check the code I changed in the last refactor for missing error handling and secrets”
  • “Run /verify every time I finish a component in this long session”
Step-by-step guide: how to use Verification LoopInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the MIT licence and an attribution note.

    verification-loop.zip
  2. 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. 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

  • Read the commands the agent is about to run and confirm they match your project before allowing them, since it executes build, test and install-related commands on your machine.
  • Add your project's real lint, test and type-check commands to the skill if they differ from the examples, otherwise the agent may guess wrong.
  • Treat the secret scan as a rough first pass and use a dedicated secret scanner before publishing code.

What's inside

  • agents/openai.yaml
  • 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.