- Home
- Skills
- Code Quality & Review
- AI Generated Code Review & Test Coverage Gate
Works with the AI tools you already use
AI Generated Code Review & Test Coverage Gate
The AI-Generated Code Review & Test-Coverage Gate is an adversarial reviewer for code written or heavily assisted by an AI agent.
$14
AI Generated Code Review & Test Coverage Gate
Example session with this skill installed
Diff: new function applyDiscount(price, qty, pct).
One test asserts applyDiscount(100,1,10) === 90.
No other tests included.
- Read your context and instructions
- Compiled the ai generated code
VERDICT: BLOCK
SUMMARY: A new pricing path ships with happy-path-only testing and an unhandled boundary that will misprice orders.
COVERAGE: Negative-quantity and zero-price branches have no test; only the 10 percent case is exercised.
SILENT CHANGES: Rounds with floor where the prior path rounded half-up, shifting totals by a cent, and this is not called out.
EDGE CASES: Discounts over 100 percent and null price are unhandled and would return negative totals or throw.
OVER-CONFIDENCE: Calls a currency helper that does not appear to exist in this module; confirm the import.
TEST QUALITY: The lone test pins a hard-coded number but would still pass if rounding changed; it does not lock the behavior.
MUST FIX BEFORE MERGE: Add tests for zero, negative, and over-100-percent inputs; confirm the helper; document or revert the rounding change.
Connects securely to your tools. The creator never sees your data.
What you get
About this skill
The AI-Generated Code Review & Test-Coverage Gate is an adversarial reviewer for code written or heavily assisted by an AI agent. You give it a pull request, a diff, or the changed files plus their tests, and it reviews the change the way a skeptical senior engineer would. It runs five passes — coverage, silent changes, edge cases, over-confidence, and test quality — and returns a PASS / REVISE / BLOCK verdict naming the untested branches, the behavior that quietly changed, and the items that must be fixed before merge.
It exists because AI code fails differently from human code: it is fluent and plausible, so reviewers trust it more than they should, and the gaps hide in branches nobody ran and assertions nobody wrote. This gate forces the boring, specific checks that catch silent regressions and happy-path-only tests. It does not run or rewrite your code — it reasons over what you provide and hands the decision back to you.
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
3 installs
Downloaded by developers to date
30-day refund
Not a fit? Get your money back
Trust & safety
Security scanned
Verified clean 3 months ago
- Passed all security checks, Safe to install