save-learnings

To save session learnings, prompt the assistant to remember the session. It extracts corrections, patterns, and knowledge, creates memory files with YAML frontmatter, and updates the MEMORY.md index while avoiding duplicates.

Persist project patterns at the end of productive sessions, after user feedback, or when prompted to save. This captures relevant knowledge and corrections so future sessions benefit from today's discoveries.

Remembering user corrections involves writing them as concise memory entries with YAML frontmatter. The system catalogs these entries and updates the MEMORY.md index automatically when topics are revisited or refined.

The memory management system avoids duplicates by checking existing entries before saving. When topics are revisited or refined, it updates existing memory entries instead of creating redundant files.

No dependencies are required to catalog session learnings. The system operates independently, creating memory files with concise YAML frontmatter and maintaining a MEMORY.md index for discovery.

Saving learnings to memory improves future assistant performance by retaining corrections and project patterns. Future sessions can recall this persisted knowledge, avoiding repeated mistakes and accelerating productivity.