Scan session transcripts and merge corrections, preferences, and decisions into persistent memory files.
npx skills add https://github.com/grandamenium/dream-skill --skill dream-grandamenium
Please help me install this Agent Skill. Skill: dream Source: https://github.com/grandamenium/dream-skill/tree/main Command: npx skills add https://github.com/grandamenium/dream-skill --skill dream-grandamenium
Memory fragmentation accumulates as sessions add context and memory files become noisy, inconsistent, and hard to navigate. This skill scans transcripts for corrections, preferences, decisions, and recurring patterns, then merges findings into persistent memory files to keep memory clean, current, and contradiction-free.
Invoke the dream command in Claude Code to start memory consolidation.
High-intent search queries and answers about installing and using this skill.
Memory consolidation for Claude Code works by scanning recent session transcripts to merge corrections, preferences, and recurring patterns into persistent memory files. It automatically triggers via the native Stop hook every 24 hours to keep memory clean and current.
Resolving contradictions in Claude Code memory involves scanning multi-session transcripts to identify conflicting preferences or decisions, then merging them into persistent files. The process converts relative dates to absolute and prunes references to nonexistent files for long-term reliability.
Yes, Claude Code memory management supports auto-triggering via the native Stop hook every 24 hours for both individual and team workflows. It auto-detects the memory system in use—native Claude Code, OpenClaw, or project-root—to determine where to read and write memory.
To keep a Claude Code MEMORY.md index under 200 lines, the consolidation process prunes redundant data and organizes memories into separate topic files. This produces a lean index while preserving all essential corrections, decisions, and recurring patterns.
You can manually run memory consolidation by invoking the dream command in Claude Code whenever session transcripts accumulate heavy fragmentation or critical preference changes. This immediately processes the four phases—Orient, Gather Signal, Consolidate, and Prune & Index—without waiting for the 24-hour auto-trigger.
Yes, Claude Code memory consolidation auto-detects OpenClaw memory systems alongside native Claude Code and project-root setups. It determines the appropriate read and write locations automatically, ensuring memory consolidation applies correctly across different detected environments.
thimsluggaConsolidate session transcripts into persistent memory with a 4-phase workflow.
mawahaConsolidate recent session learnings into organized Markdown topic files.
liujiayi3703Manage persistent memory and observation feeds for OpenClaw agents via the claude-mem plugin.
AlvaroLaraFFAudit memory entries for age, vagueness, and data leaks.
