Independent Verifier Pass
When implementation claims done, a separate verifier pass runs build, lint, and tests with no access to the implementer's rationale.
Kickoff prompt
Start the "Independent Verifier Pass" loop.
Goal: build, lint, and tests pass under independent verification
Max iterations: 8
Between iterations run: npm run build && npm run lint && npm test
Exit when: all verifier commands exit 0
Step 1: Run build, lint, and tests as a verifier. Trust only command output, not prior claims.
Self-pace this loop. After each iteration, run the check command, read the output, and only continue if the exit condition is not met. Stop when the exit condition passes or max iterations is reached. Give a short status update each pass.Basic information
Category
Testing
Goal
build, lint, and tests pass under independent verification
Max iterations
8
Check command
npm run build && npm run lint && npm test
Exit when
all verifier commands exit 0
Supported agents
Claude Code
Steps
- 1
Run verifier checks
As an independent verifier, run build, lint, and tests. Do not assume the prior pass was correct.
npm run build && npm run lint && npm test - 2
Report gaps
List every failing check with file paths and error excerpts. No self-reporting — only command output counts.
- 3
Fix or hand back
If failures exist, fix them or produce a concise failure report for the implementer to continue.
More Testing loops
Build Until Green
Run the production build, fix compile and bundling errors, and loop until the build succeeds.
Coverage Until Threshold
Add focused tests until coverage meets your threshold (e.g. 80%), without changing production behavior unnecessarily.
Pre-Commit Guard
Hook that runs tests before git commit commands to block commits when the suite is red.
Post-Edit Test Guard
Hook-based loop that runs related tests after file edits to catch regressions early.
Format Until Clean
Run the formatter, fix any remaining style issues, and loop until the codebase formats cleanly with no diffs.
Flaky Test Triage
Run failing tests repeatedly, classify each failure as flaky or real, and fix only confirmed regressions.