What problem does it solve? When an AI coding agent finishes a work session, the resulting changes still need to be reviewed, grouped, and committed to git. This Skill automates that final step while ensuring commit messages contain no AI attribution, co-author lines, or generated-with footers, so the git history reads as if the user wrote it. ## Core Features & Use Cases - Change Analysis: Runs git status and git diff to understand exactly what changed during the session before committing. - Logical Commit Grouping: Splits changes into focused, atomic commits using targeted git add of specific files rather than blanket staging. - Clean Commit Messages: Writes imperative-mood messages focused on why changes were made, with no co-author or Codex attribution lines. - Use Case: After a fastflow workflow finishes implementing a feature, invoke this Skill to turn the working-tree changes into one or more well-described commits ready for review or a pull request. ## Quick Start Ask the agent to run the ff_commit command to commit the changes made in this session.