- Home
- Skills
- Code Quality & Review
- AI Agent Output Verification Gate™
Works with the AI tools you already use
AI Agent Output Verification Gate™
Verify AI-generated code against task requirements, evidence, tests, and merge-readiness criteria.
$14.99
AI Agent Output Verification Gate™
Example session with this skill installed
Verify the AI agent's implementation of this task before I merge it.
ORIGINAL TASK
Add password-reset token expiration after 15 minutes.
ACCEPTANCE CRITERIA
- Reset tokens must expire after exactly 15 minutes.
- Expired tokens must be rejected.
- A useful error should be returned for expired tokens.
- Add tests covering token expiration.
- Existing password-reset functionality must continue working.
The AI agent says the task is complete.
Inspect the current git diff, relevant source files, and existing tests.
Do not rewrite the code yet.
Give me a PASS / REVIEW / BLOCK decision based on evidence.
- Read your context and instructions
- Compiled the ai agent output
AI AGENT OUTPUT VERIFICATION
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
FINAL DECISION: BLOCK
COMPLETION SCORE: 72%
RISK: HIGH
TASK
Add password-reset token expiration after 15 minutes.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
REQUIREMENT VERIFICATION
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
[PASS] 1. Token expires after exactly 15 minutes
Evidence
src/auth/reset-token.ts
Expiry value is configured to 900 seconds.
[PASS] 2. Expired tokens are rejected
Evidence
src/auth/reset.service.ts
Expired-token validation is present before reset execution.
[PASS] 3. Expired-token error is returned
Evidence
src/auth/errors.ts
Existing TOKEN_EXPIRED error is returned by the reset flow.
[FAIL] 4. Expiration tests are incomplete
Found
- Token creation test
Missing
- Token exactly at expiry boundary
- Expired-token rejection test
[WARN] 5. Existing functionality
Existing password-reset tests pass, but the new expiration path is not covered by an integration test.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
EVIDENCE SUMMARY
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
VERIFIED REQUIREMENTS: 3/5
PARTIALLY VERIFIED: 1
UNVERIFIED: 1
MISSING VERIFICATION
- Expired-token integration test
- Expiry-boundary test
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
REGRESSION RISK
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
HIGH
Reason
The implementation changes authentication behavior, but the new failure path lacks sufficient automated coverage.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
REQUIRED BEFORE MERGE
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
- Add an expired-token integration test.
- Add a boundary test for the 15-minute expiry.
- Re-run the password-reset test suite.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
VERIFICATION CRITERIA
━━━━━━━━━━━━━━━━━━━━━━━━━━━━
PASS only when
✓ Valid token works before expiry
✓ Token is rejected after expiry
✓ Expiry boundary behaves as intended
✓ Existing password-reset flow remains functional
✓ Relevant tests pass
FINAL
BLOCK MERGE
Connects securely to your tools. The creator never sees your data.
What you get
About this skill
AI coding agents can produce working-looking changes while still missing requirements, edge cases, tests, or acceptance criteria.
AI Agent Output Verification Gate™ is a structured verification workflow that checks whether an AI agent actually completed the requested task—not simply whether the generated code looks reasonable.
Provide the original task or acceptance criteria together with the AI agent's changes, diff, relevant files, or test results. The skill traces each important requirement to repository evidence, checks affected behavior, identifies missing verification, evaluates regression risk, and produces a clear decision.
You receive
Requirement-by-requirement verification Evidence supporting completed claims Unverified or partially completed work Missing test coverage Regression-risk assessment Required fixes before merge Verification steps for unresolved items Final PASS / REVIEW / BLOCK decision Completion score showing how much of the requested work is actually verified
Designed for developers and teams using Claude Code, Codex CLI, Cursor, Copilot, Gemini CLI, and other AI coding agents.
This is not a generic code review. It verifies the AI agent's claimed work against the original task and available repository evidence.
How to install
Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.
- 1
Download the ZIP
Free skills download straight away. Paid skills unlock right after purchase.
- 2
Unzip into your skills folder
Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.
- 3
Ask your agent to use it
Restart the agent if it was already running. It picks the skill up automatically - no config needed.
Skills folder by agent
Click the path to copy it. Create the folder if it does not exist yet.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
30-day refund
Not a fit? Get your money back
Trust & safety
Security scanned
Verified clean 1 month ago
- Passed all security checks, Safe to install