Verification Loop Skill

A comprehensive verification system for Claude Code sessions.

When to Use

Invoke this skill:

• After completing a feature or significant code change
• Before creating a PR
• When you want to ensure quality gates pass
• After refactoring

Verification Phases

Pick the right command for each phase from the project's own configuration (package.json scripts, Makefile, CI workflow, etc.). The commands below are examples, not prescriptions. Summarize long output instead of pasting it verbatim.

Phase 1: Build Verification

Check that the project builds. Examples: npm run build, cargo check, go build ./..., shellcheck -S warning *.sh.

If the build fails, STOP and fix before continuing.

Phase 2: Type Check

Run the project's type checker if it has one. Examples: npx tsc --noEmit (TypeScript), pyright . or mypy . (Python).

Report all type errors. Fix critical ones before continuing.

Phase 3: Lint Check

Run the project's linter. Examples: npm run lint, ruff check ., shellcheck.

Phase 4: Test Suite

Run the project's test suite, with coverage if available. Examples: npm test -- --coverage, pytest --cov, bash tests/run-unit-tests.sh.

Report:

• Total tests: X
• Passed: X
• Failed: X
• Coverage: X% (against the project's target)

Phase 5: Security Scan

Check for leaked secrets in the changes:

• If a dedicated scanner is available (gitleaks, trufflehog), use it.
• Otherwise, follow the bundled security-review skill.
• At minimum, review the added lines of git diff for credentials, tokens, and keys. Do not report PASS based on a repository-wide grep.

Also check for leftover debug output (e.g., console.log, print) in the changed files.

Phase 6: Diff Review

# Show what changed
git diff --stat
git diff HEAD~1 --name-only

Review each changed file for:

• Unintended changes
• Missing error handling
• Potential edge cases

Output Format

After running all phases, produce a verification report:

VERIFICATION REPORT
==================

Build:     [PASS/FAIL]
Types:     [PASS/FAIL] (X errors)
Lint:      [PASS/FAIL] (X warnings)
Tests:     [PASS/FAIL] (X/Y passed, Z% coverage)
Security:  [PASS/FAIL] (X issues)
Diff:      [X files changed]

Overall:   [READY/NOT READY] for PR

Issues to Fix:
1. ...
2. ...

Integration with Hooks

This skill complements PostToolUse hooks but provides deeper verification. Hooks catch issues immediately; this skill provides comprehensive review. If you need periodic automated re-verification, use a hook or the /loop command; instructions in this skill body cannot schedule themselves.