What problem does it solve?
Extract reusable patterns, decisions, and failure learnings from completed work into documentation that compounds knowledge over time. This process ensures organizational memory is built from real-world outcomes rather than isolated code changes.
Core Features & Use Cases
- Durable learning extraction: capture patterns, decisions, and failures from a completed task into a structured knowledge artifact.
- Session handoff and post-merge compounding: provide compact handoff data for ongoing work and write dated learnings after a merge or abandonment.
- Dream consolidation: perform on-demand Codex-based consolidation passes over accumulated learnings to improve future work.
- Knowledge linking: tie learnings back to source tasks and artifacts for traceability.
Quick Start
Process a completed feature to generate a durable learnings document and a compact handoff payload.