How does Memory Fabric handle duplicate information from different sources?

It employs a deduplication engine that identifies results with over 85% text similarity, merging their metadata and ranking the most authoritative source.

What is the primary source of memory in Memory Fabric?

Memory Fabric uses a Knowledge Graph (via MCP memory servers) as its primary, always-available data source for entity relationships and context.

Can I query complex relationships with Memory Fabric?

Yes, the skill supports multi-hop graph traversal, allowing Claude to navigate through multiple layers of connected entities to find relevant context.

How does the cross-reference boost work?

When a semantic search result from mem0 mentions an entity that already exists in the local Knowledge Graph, the relevance score is boosted by 1.2x to reflect its verified importance.

Is a mem0 API key required to use this skill?

No, mem0 is an optional enhancement for semantic cloud search. Memory Fabric is fully functional using only the local Knowledge Graph if mem0 is not configured.

Memory Fabric

Name: Memory Fabric
Author: yonatangross

byyonatangross

•

Gestión de Bases de Datos

Orchestrates graph-based and semantic memory systems to provide unified, high-context information retrieval for AI development.

Memory Fabric is an advanced orchestration layer for Claude Code that harmonizes local knowledge graphs with cloud-based semantic search. By prioritizing a graph-first architecture while optionally integrating mem0, it ensures that Claude has access to the most relevant, deduplicated, and cross-referenced project context. This skill is essential for managing complex software projects where maintaining a coherent history of architectural decisions, entity relationships, and multi-hop knowledge traversal is critical for accurate AI assistance and RAG performance.

Características Principales

01Unified ranking system combining recency, relevance, and source authority

0269 GitHub stars

03Graph-first memory orchestration using primary MCP memory servers

04Automated deduplication engine with >85% similarity detection

05Multi-hop graph traversal for complex entity relationship mapping

06Cross-referencing between semantic cloud storage and local knowledge graphs

Casos de Uso

01Mapping relationships between project entities like agents, technologies, and patterns

02Retrieving historical architectural decisions and rationales across large codebases

03Building a unified context layer for highly accurate RAG (Retrieval-Augmented Generation)

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yonatangross/orchestkit memory-fabric

For use in Claude.ai and ChatGPT

Download Skill