What is HNSW vector search?

HNSW (Hierarchical Navigable Small World) is an algorithm for approximate nearest neighbor search, providing extremely fast semantic lookups (up to 12,500x faster) for finding similar patterns in memory.

Can I share memory between different Claude agents?

Yes, the system supports hierarchical namespaces, allowing different agents to store, retrieve, and share knowledge across the same persistent database.

Is the stored data persistent and portable?

Yes, the skill provides persistent storage via AgentDB and includes commands for exporting memory to JSON and creating backups to external storage.

How does this improve AI coding efficiency?

By storing successful code patterns and learnings, Claude can avoid repeating mistakes and instantly recall optimized solutions from previous sessions.

Memory Management

Name: Memory Management
Author: nbossn

bynbossn

0•

データベース管理

Manages persistent AgentDB storage with HNSW vector search for high-speed semantic pattern retrieval and cross-agent knowledge sharing.

The Memory Management skill integrates an advanced AgentDB system with HNSW (Hierarchical Navigable Small World) vector search to provide Claude with long-term persistent memory. It enables agents to store successful implementation patterns, perform semantic searches across previous tasks, and share knowledge across different sessions with retrieval speeds up to 12,500x faster than traditional methods. This skill is essential for developers looking to build a reusable codebase knowledge base, learn from past iterations, and maintain consistency in complex software development workflows.

主な機能

010 GitHub stars

02High-speed HNSW vector search for semantic pattern retrieval

03Persistent AgentDB storage for long-term knowledge retention

04Cross-agent knowledge sharing via hierarchical namespaces

05Memory optimization tools including consolidation and backups

06Detailed usage statistics and JSON export capabilities

ユースケース

01Storing and retrieving successful code implementation patterns

02Building a project-specific knowledge base for AI agents

03Semantic lookup of past solutions to avoid redundant work

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nbossn/claude-code-base memory-management

For use in Claude.ai and ChatGPT

主な機能

010 GitHub stars

02High-speed HNSW vector search for semantic pattern retrieval

03Persistent AgentDB storage for long-term knowledge retention

04Cross-agent knowledge sharing via hierarchical namespaces

05Memory optimization tools including consolidation and backups

06Detailed usage statistics and JSON export capabilities

ユースケース

01Storing and retrieving successful code implementation patterns

02Building a project-specific knowledge base for AI agents

03Semantic lookup of past solutions to avoid redundant work

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nbossn/claude-code-base memory-management

For use in Claude.ai and ChatGPT