infinite-memory-v2
by mhndayesh
Notes & PKMBrief
High-precision memory with 100% recall accuracy for long contexts.
Description
What This Skill Does
RAG-based memory engine that stores and retrieves facts across long contexts using a parallel LMS architecture. Exposes two operations: ingesting text into a persistent store and querying it by natural language. Designed for high recall accuracy over extended history.
Retrieves specific facts from large historical datasets where a standard context window would truncate or lose older information entirely.
When to Use It
- Recalling project decisions made weeks or months ago
- Storing meeting notes and retrieving them by topic later
- Querying past research findings without re-reading source documents
- Persisting key facts between long agent conversations
- Building a searchable personal knowledge base from unstructured text
Example Workflow
User asks: What did we decide about the API rate limits last sprint?
- 1
Runs recall_facts with the query string against the historical database
- 2
Retrieves top matching fact entries ranked by relevance
- 3
Returns the matched content with source attribution
Stored fact: 'API rate limit set to 500 req/min per user, agreed 2024-11-14'
Installation
clawhub install mhndayesh/infinite-memory-v2Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.