ai-code-review

1# AI Code Review Skill
2 
3Human code fails where the human got tired or didn't know; AI code fails where *plausibility diverged from correctness* — and it fails fluently, with confident naming, clean formatting, and tests that pass without testing anything. Reviewing it with human-code instincts ("looks careful, probably is careful") is how the new bug class ships. This skill reviews for the failure modes that are characteristically AI.
4 
5> **Not quite this?** Use `code-review-guide` when the code was written by people and needs a general review.
6 
7## What This Skill Produces
8 
9- A **review of the change** organised by AI-characteristic risk, each finding with file/line and severity
10- **Verification steps** the reviewer must actually run (not read) per risk class
11- A **team checklist** for AI-authored PRs, calibrated to this codebase
12 
13## Required Inputs
14 
15Ask for (if not already provided):
16- **The diff or PR** (or the files changed)
17- **Provenance honestly**: fully agent-written, human-piloted, or mixed — and whether the *author reviewed it themselves* before requesting review
18- **The codebase context**: existing conventions/utilities the AI may not have known, and what the change claims to do
19- **Test infrastructure**: what CI actually runs (the AI may have written tests CI never executes)
20 
21## The AI-Characteristic Failure Modes
22 
23Review in this order — most damaging first:
24…