AI Code Review Toolkit: PR Checklist & Risk Tracker for Claude Code, Cursor & Copilot Teams

A PR review toolkit built around six failure patterns that Claude Code, Cursor, and GitHub Copilot repeatedly introduce: hardcoded secrets, missing authorization checks, mock-heavy tests, hallucinated API calls, silent edge-case failures, and hidden scope creep.

What's included:

— 16-page PDF: the 6 Failure Families explained, a 5-Minute Triage Checklist, a 45-item Full Review Checklist across six risk categories, a 1-5 Risk Scoring Rubric, solo + team reviewer workflows, 6 ready-to-paste PR comment templates, and tool-specific notes for Claude Code, Cursor, and Copilot

— PR Risk Tracker spreadsheet (.xlsx, Google Sheets & Excel compatible) with a Dashboard tab that surfaces which failure pattern and which AI tool produce the most rework

Who it's for: solo developers and small engineering teams (2-15 people) who've adopted AI coding agents for a meaningful share of their code and don't yet have a formal process for reviewing what those agents produce before it merges.