diagnose-agent

You can profile an AI agent's strengths and weaknesses by analyzing git history, ownership patterns, and learnings evolution to produce an evidence-based assessment of durable skills and knowledge gaps.

Agent profiling from learnings history and git signals is the process of mining commit data and past learned behavior to surface fix-after-feature signals, churn, and sparse knowledge areas for performance review.

Diagnose a team agent by evaluating its historical learnings and git commit patterns to determine whether it needs foundational gaps filled or adversarial edge-case practice before assigning harder work.

Yes, you can use git history to identify knowledge gaps by reading an agent's learnings files to find churned false positives and sparse knowledge areas, producing a ranked struggle profile for downstream use.

No, agent profiling requires only read-only access to team configuration, learnings files, and git history to produce a pipe-formatted struggle profile without modifying the repository.

The best way to assess agent performance using commit ownership patterns is to mine git signals for fix-after-feature trends and churn, outputting a ranked, pipe-friendly struggle profile for downstream challenge generation.