What problem does it solve? AI coding assistants often introduce verbose artifacts into codebases: unnecessary comments, defensive checks, try/catch blocks, and type casts that a human developer would never write. This Skill cleans up those artifacts by reviewing the diff against main and stripping out the slop before merge. ## Core Features & Use Cases - Diff-Based Review: Checks the git diff against main to identify all AI-generated slop introduced in the current branch. - Selective Comment Cleanup: Removes redundant comments while preserving comments that explain reasoning, context, trade-offs, or non-obvious knowledge. - Defensive Code Removal: Eliminates abnormal defensive checks, try/catch blocks, and casts to any that are inconsistent with the surrounding codebase style. - Use Case: After an AI assistant implements a feature branch, run this Skill before opening a pull request to ensure the diff matches the codebase's human-written style and contains no leftover AI verbosity. ## Quick Start Review the diff of my current branch against main and remove all AI-generated slop, then summarize the changes in Japanese.