What problem does it solve? When AI coding agents write your code, the reasoning behind decisions is lost once the session ends. This Skill recovers that context by linking commits to their originating agent sessions, so you can answer why code exists, who or what wrote it, and whether an agent has solved a similar problem before. ## Core Features & Use Cases - Code Provenance: Use review trace blame to map lines of code to commits, then pull the agent sessions behind them via Agent-Session commit trailers. - Trace Search: Search normalized session transcripts with FFF and inspect exact events using review trace show with session, trace, and event locators. - Topic Research: Page through all published sessions with review trace sessions to find whether an agent has previously handled a problem. - Use Case: While debugging a confusing function, run blame on its lines, pull the linked session, and read the agent's original reasoning before deciding whether to change the code. ## Quick Start Ask the agent to find out why a specific piece of code exists by tracing it back to the agent session that wrote it.
