Consolidate session logs into durable memory files and clean stale entries.
npx skills add https://github.com/stewnight/rem-sleep-skill --skill rem-sleep
Please help me install this Agent Skill. Skill: rem-sleep Source: https://github.com/stewnight/rem-sleep-skill/tree/main Command: npx skills add https://github.com/stewnight/rem-sleep-skill --skill rem-sleep
This Skill addresses the challenge of accumulating session logs that are expensive to re-read, leading to important insights being buried and memory loss after context compaction or restarts.
Run the REM sleep skill to consolidate recent session logs into durable memory.
High-intent search queries and answers about installing and using this skill.
Memory consolidation processes raw session logs to extract significant patterns and preserve crucial decisions, preventing knowledge loss after context compaction or restarts. This optimization ensures important insights are not buried in expensive log accumulation.
Preventing data loss from accumulating session logs is achieved through periodic memory maintenance that extracts significant patterns from raw logs and stores them in durable memory files, ensuring crucial decisions and learned facts are preserved.
Memory defragmentation works by reviewing and cleaning up long-term memory files, removing stale entries and duplicates, and compressing verbose information to optimize long-term knowledge storage for AI agents.
Yes, you need shell scripting for log searching and parsing to process session logs for memory consolidation. Optional integration with tools like Repo Prompt can provide enhanced semantic search capabilities.
You should run memory maintenance periodically to ensure crucial decisions, learned facts, and important preferences are preserved in long-term memory, preventing the loss of valuable knowledge before log accumulation becomes expensive to re-read.
DavidROliverBAClear the current AI session and generate a context handoff summary.
X-McKaySearch hierarchical memory tiers and return relevance-ranked results.
kyopark2014Stores and retrieves agent memory via Markdown files and semantic search.
hbvg234Back up Claude transcripts before context compaction with JSON metadata.
