learnself

Agent self-learning works by scanning prior session transcripts during idle time, extracting repeating patterns and lessons, and persisting them into a writable learning memory location for future performance improvement.

Multi-agent handoff transfers pending work by packaging context including session state, task details, files, and progress, then tracking pending, accepted, completed, or failed states across different agents.

Yes, session retrospective learning applies to idle-time scenarios where agents scan execution history, extract lessons automatically, and maintain workflow continuity by resuming unfinished tasks reliably across different agents.

Session end work clearing consolidates unfinished tasks, prioritizes them, creates handoffs for tasks that must continue, records status, and confirms to the human that the session is properly wrapped up.

Multi-agent gang integration supports per-gang memory separation for isolated learning, while offering optional sharing of learnings across gangs to enable team-level performance improvement and collaborative knowledge extraction.

Agent memory extraction and handoff packaging require access to session transcripts for pattern learning and a writable learning memory location to persist extracted knowledge and track handoff states reliably.