MCPgee
MCPgee
MCPgeeMCPgee

Knowledge Base

v0.2.1•Knowledge & Memory•stable

A zero-dependency MCP server that provides a persistent personal knowledge base for Claude using local JSON file storage. It enables users to manage notes through tools for adding, searching, and indexing content built entirely with Node.js built-ins

knowledge-base-mcp-servermcpai-integration
Share:
47
Stars
0
Downloads
0
Weekly
0/5
View on GitHub

What is Knowledge Base?

Knowledge Base is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to zero-dependency mcp server that provides a persistent personal knowledge base for claude using local json file storage. it enables users to manage notes through tools for adding, searching, and indexi...

A zero-dependency MCP server that provides a persistent personal knowledge base for Claude using local JSON file storage. It enables users to manage notes through tools for adding, searching, and indexing content built entirely with Node.js built-ins

This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • A zero-dependency MCP server that provides a persistent pers

Use Cases

Create and manage a persistent personal knowledge base with Claude
Search and index notes using JSON file storage
jeanibarz

Maintainer

LicenseThe Unlicense
Languagetypescript
Versionv0.2.1
UpdatedMay 22, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux
View SourceBrowse All Servers

Installation

NPM

npx -y @jeanibarz/knowledge-base-mcp-server

Manual Installation

npx -y @jeanibarz/knowledge-base-mcp-server

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Knowledge Base

The Knowledge Base MCP Server is a zero-dependency personal knowledge management system for Claude and other MCP clients, backed by local JSON file storage and FAISS vector indexing. It lets you build a persistent knowledge base that Claude can add to, semantically search, and query with LLM-generated answers with citations — all stored locally on your machine. The server supports multiple embedding backends (Ollama, OpenAI, HuggingFace) and exposes tools for document ingestion, retrieval, indexing, and statistics.

Prerequisites

  • Node.js 18 or later (for npx execution)
  • An embedding backend: Ollama running locally (recommended, free) or an OpenAI API key
  • An MCP-compatible client such as Claude Desktop
  • Sufficient disk space for FAISS indexes and document storage (default location: ~/knowledge_bases)
1

Set up an embedding backend

Install Ollama (https://ollama.com) and pull an embedding model, or prepare an OpenAI API key. Ollama is recommended for privacy and zero cost.

# With Ollama (recommended)
ollama pull dengcao/Qwen3-Embedding-0.6B:Q8_0
2

Test the server runs

Run the server once via npx to confirm it starts correctly.

EMBEDDING_PROVIDER=ollama OLLAMA_BASE_URL=http://localhost:11434 npx -y @jeanibarz/knowledge-base-mcp-server@latest
3

Add to your MCP client configuration

Edit your claude_desktop_config.json to register the knowledge base server with your chosen embedding backend configuration.

{
  "mcpServers": {
    "knowledge-base": {
      "command": "npx",
      "args": ["-y", "@jeanibarz/knowledge-base-mcp-server@latest"],
      "env": {
        "EMBEDDING_PROVIDER": "ollama",
        "OLLAMA_BASE_URL": "http://localhost:11434",
        "OLLAMA_MODEL": "dengcao/Qwen3-Embedding-0.6B:Q8_0",
        "KNOWLEDGE_BASES_ROOT_DIR": "/Users/yourname/knowledge_bases",
        "FAISS_INDEX_PATH": "/Users/yourname/knowledge_bases/.faiss"
      }
    }
  }
}
4

Restart Claude Desktop

Fully quit and reopen Claude Desktop so it loads the new server configuration and connects to the knowledge base.

5

Add your first document

Ask Claude to add a document to your knowledge base. You can also use the CLI's 'kb remember' command to write notes directly.

6

Search and query your knowledge base

Once documents are added, ask Claude to search or answer questions from your knowledge base. The server uses semantic search (FAISS dense retrieval) and can provide LLM-generated answers with citations.

Knowledge Base Examples

Client configuration

Add this to claude_desktop_config.json to use the knowledge base server with Ollama embeddings. For OpenAI embeddings, replace the Ollama env vars with EMBEDDING_PROVIDER=openai and OPENAI_API_KEY.

{
  "mcpServers": {
    "knowledge-base": {
      "command": "npx",
      "args": ["-y", "@jeanibarz/knowledge-base-mcp-server@latest"],
      "env": {
        "EMBEDDING_PROVIDER": "ollama",
        "OLLAMA_BASE_URL": "http://localhost:11434",
        "OLLAMA_MODEL": "dengcao/Qwen3-Embedding-0.6B:Q8_0",
        "KNOWLEDGE_BASES_ROOT_DIR": "/Users/yourname/knowledge_bases",
        "FAISS_INDEX_PATH": "/Users/yourname/knowledge_bases/.faiss"
      }
    }
  }
}

Prompts to try

Use these prompts in Claude Desktop after connecting the knowledge base server.

- "Add this note to my knowledge base: [paste text here]"
- "Search my knowledge base for information about project planning"
- "What do I know about Kubernetes based on my notes?"
- "List all my knowledge bases"
- "Show statistics for my main knowledge base"
- "Import this URL into my knowledge base: https://example.com/article"

Troubleshooting Knowledge Base

INDEX_NOT_INITIALIZED error when searching

You need to add at least one document before searching. The FAISS index is built on first ingestion. Add a document via Claude or using 'kb remember --stdin --yes' from the CLI, then search.

Embedding model fails with context length error

Your chosen embedding model has too small a context window (minimum ~500 tokens required). Avoid models like all-minilm which have 256 token context. Use the recommended 'dengcao/Qwen3-Embedding-0.6B:Q8_0' model or 'text-embedding-3-small' from OpenAI.

PROVIDER_AUTH error — OpenAI authentication fails

Ensure OPENAI_API_KEY is set correctly in the env block of your MCP config. The key must have access to the Embeddings API. Run 'kb doctor' from the CLI to get a health check report on all configured providers.

Frequently Asked Questions about Knowledge Base

What is Knowledge Base?

Knowledge Base is a Model Context Protocol (MCP) server that zero-dependency mcp server that provides a persistent personal knowledge base for claude using local json file storage. it enables users to manage notes through tools for adding, searching, and indexing content built entirely with node.js built-ins It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Knowledge Base?

Install via npm with the command: npx -y @jeanibarz/knowledge-base-mcp-server. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with Knowledge Base?

Knowledge Base works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Knowledge Base free to use?

Yes, Knowledge Base is open source and available under the The Unlicense license. You can use it freely in both personal and commercial projects.

Learn More About MCP Servers

Getting Started with MCP

Set up your first MCP server in minutes

MCP Setup Guide

Configure MCP in Claude, Cursor & VS Code

All MCP Tutorials

18+ hands-on guides for developers

MCP FAQ

40+ answers about Model Context Protocol

Knowledge Base Alternatives — Similar Knowledge & Memory Servers

Looking for alternatives to Knowledge Base? Here are other popular knowledge & memory servers you can use with Claude, Cursor, and VS Code.

MemPalace

★ 52.6k

A local AI memory system that stores all conversations verbatim and organizes them into navigable structures. It provides 19 MCP tools for AI assistants to search and retrieve past decisions, debugging sessions, and architecture debates automatically

Kratos

★ 25.7k

🏛️ Memory System for AI Coding Tools - Never explain your codebase again. MCP server with perfect project isolation, 95.8% context accuracy, and the Four Pillars Framework.

Context Mode

★ 15.4k

An MCP server that preserves LLM context by intercepting large data outputs and returning only concise summaries or relevant sections. It enables efficient sandboxed code execution, file processing, and documentation indexing across multiple programm

Memu

★ 13.7k

Memory for 24/7 proactive agents like OpenClaw.

MemOS

★ 9.3k

MemOS (Memory Operating System) is a memory management operating system designed for AI applications. Its goal is: to enable your AI system to have long-term memory like a human, not only remembering what users have said but also actively invoking, u

Everos

★ 5.4k

Build, evaluate, and integrate long-term memory for self-evolving agents.

Browse More Knowledge & Memory MCP Servers

Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Knowledge & MemoryBrowse All Servers

Quick Config Preview

{ "mcpServers": { "knowledge-base-mcp-server": { "command": "npx", "args": ["-y", "@jeanibarz/knowledge-base-mcp-server"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Knowledge Base?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides
Explore All ServersRead Our Guides