consolidate-memory

Name: consolidate-memory
Availability: InStock
Author: karan

Consolidate auto-memory directories by merging overlapping topic files and rewriting MEMORY.md.

16|2|Updated Feb 12, 2026

Search Tags:#knowledge management #deduplication #user preferences #memory consolidation #index maintenance #time normalization

Ready to run

npx skills add https://github.com/thtskaran/claude-skills --skill consolidate-memory-thtskaran

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: consolidate-memory
Source: https://github.com/thtskaran/claude-skills/tree/main/consolidate-memory
Command: npx skills add https://github.com/thtskaran/claude-skills --skill consolidate-memory-thtskaran

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents your memory index from drifting out of date by consolidating overlapping and stale memory files so future sessions can quickly understand the user, priorities, and working preferences without re-asking.

Core Features & Use Cases

Duplicate and overlap merging: Combines memory topics that describe the same person, project, or preference.
Stale-pruning and durable retention: Separates durable preferences and relationships from dated, already-finished tasks and deadlines.
Time reference normalization: Converts relative time phrases into absolute dates so the memories remain readable long-term.
Index cleanup with size constraints: Rewrites MEMORY.md to stay under 200 lines and ~25KB with compact one-line entries.

Quick Start

Ask the AI to run the Memory Consolidation workflow by reviewing MEMORY.md, scanning each topic file for overlap or staleness, then merging, updating dates, pruning, and finally rewriting the index under the specified limits.

Frequently Asked Questions about consolidate-memory

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I deduplicate overlapping topics in my AI memory files?▼

Memory consolidation deduplicates overlapping topics by scanning topic markdown files, identifying duplicate descriptions of people or projects, merging the information, and rewriting a concise index to prevent memory drift.

What is the best way to prune stale items from a user memory index?▼

Pruning stale items involves separating durable user preferences from dated, completed tasks, then retiring the outdated entries so future sessions load accurate context without re-asking questions.

How do I normalize relative time references in markdown memory files?▼

Time normalization converts relative time phrases into absolute dates during memory consolidation, ensuring long-term readability so saved memories remain accurate regardless of when they are accessed.

Can I use memory consolidation on a large MEMORY.md index file?▼

Yes, memory consolidation rewrites the MEMORY.md index to stay under 200 lines and approximately 25KB, using compact one-line entries to maintain a lean and quickly scannable user profile.

When should I run a memory consolidation workflow on my topic files?▼

You should run memory consolidation as post-session maintenance when your auto-memory directory accumulates overlapping topic files, stale deadlines, or relative time references that degrade future session context.