close-session

To prevent context loss during session handoff, you need to audit daily patterns and generate a concise summary of completed work and next steps. This ensures the AI starts the next interaction with full continuity and updated persistent memory files.

Managing AI memory bloat requires implementing strict pruning and validation logic on your knowledge base entries. This maintains memory health by automatically enforcing size constraints on persistent files like MEMORY.md to prevent context overload.

Promoting recurring user patterns to long-term knowledge management rules involves identifying repeated behaviors across sessions and promoting them to formal rules or concept articles. This systematically captures workflow patterns into persistent memory structures.

You do not need a specific framework to automate session handoff documentation, as this process requires no external dependencies. It operates by triggering a closing ritual that updates persistent memory files directly within your existing software engineering environment.

You should run a session closing ritual for context optimization at the end of your workday or when an interaction is complete. This triggers the lifecycle management of knowledge base entries and ensures proper session handoff documentation for future interactions.