retrospective-agent

Automating retrospective analysis requires reading track metadata, commits, and spec files to extract execution learnings. This agent processes those inputs to generate actionable insights, logging new patterns and errors into a shared knowledge base for continuous improvement.

Automated post-track analysis captures errors by evaluating track plans and commit history against execution metadata. It identifies recurring failures and improvement opportunities, writing structured error logs and actionable retrospective documents to inform future development cycles.

Automated retrospective analysis applies directly to QA sprints and evaluation loops. By reading track plans and execution metadata, it extracts patterns and errors from these cycles to support continuous improvement and update knowledge bases.

The best way to extract learnings is to analyze track commits and specifications automatically. This approach aggregates execution patterns and errors, updating knowledge base files to continuously improve future executions without manual review overhead.

Running an automated retrospective requires track plan, metadata, commits, and spec files. The agent reads these specific inputs to generate a retrospective document and update the knowledge base with new patterns and error logs.