Does this support multi-agent systems?

Absolutely. It enables a unified query interface and cross-agent memory sharing, allowing multiple agents to learn from and access the same optimized memory pool.

What is HNSW indexing in the context of this skill?

Hierarchical Navigable Small World (HNSW) is a state-of-the-art algorithm for approximate nearest neighbor search, providing extreme speedups for semantic vector retrieval within Claude's memory.

Can I migrate my existing Claude Code data?

Yes, the skill includes dedicated migration strategies to move data from legacy SQLite, Markdown files, and other backends directly into the unified AgentDB.

How much faster is the search performance really?

Depending on the dataset size and query type, this unification achieves search speed improvements ranging from 150x to 12,500x compared to legacy backends.

Is this skill backward compatible?

Yes, it implements a unified memory service that maintains existing API interfaces while significantly upgrading the underlying storage and indexing architecture.

V3 Memory Unification

Name: V3 Memory Unification
Author: nbossn

bynbossn

0•

データベース管理

Consolidates multiple legacy memory systems into a high-performance AgentDB backend with HNSW vector search.

V3 Memory Unification is an advanced architectural upgrade for Claude Code that streamlines disparate storage systems—including SQLite, Markdown, and distributed systems—into a unified AgentDB with HNSW indexing. This skill implements ADR-006 and ADR-009 to deliver unprecedented search performance, achieving up to 12,500x improvement while maintaining full backward compatibility. It is essential for complex multi-agent environments requiring low-latency semantic retrieval, cross-agent memory sharing, and SONA learning pattern integration for more intelligent and responsive agent behavior.

主な機能

010 GitHub stars

02HNSW vector search with 150x-12,500x speed improvements

03Real-time cross-agent memory sharing and synchronization

04SONA integration for advanced learning pattern storage and retrieval

05Unified AgentDB backend replacing 6+ legacy memory systems

06Automated migration pathways for SQLite and Markdown data

ユースケース

01Migrating legacy file-based and SQL memories to a high-performance vector database

02Implementing semantic search across massive datasets with sub-100ms latency

03Optimizing multi-agent workflows where shared context and low-latency retrieval are critical

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nbossn/claude-code-base v3-memory-unification

For use in Claude.ai and ChatGPT

主な機能

010 GitHub stars

02HNSW vector search with 150x-12,500x speed improvements

03Real-time cross-agent memory sharing and synchronization

04SONA integration for advanced learning pattern storage and retrieval

05Unified AgentDB backend replacing 6+ legacy memory systems

06Automated migration pathways for SQLite and Markdown data

ユースケース

01Migrating legacy file-based and SQL memories to a high-performance vector database

02Implementing semantic search across massive datasets with sub-100ms latency

03Optimizing multi-agent workflows where shared context and low-latency retrieval are critical

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nbossn/claude-code-base v3-memory-unification

For use in Claude.ai and ChatGPT