MCPgee
MCPgee
MCPgeeMCPgee

Local Faiss

v1.0.0•Knowledge & Memory•stable

Local FAISS vector store as an MCP server – Agent Memory, drop-in local semantic search for Claude / Copilot / Agents.

agentic-aiai-agentsai-toolsfaissllm-tools
Share:
30
Stars
0
Downloads
0
Weekly
0/5
View on GitHub

What is Local Faiss?

Local Faiss is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to local faiss vector store as an mcp server – agent memory, drop-in local semantic search for claude / copilot / agents.

Local FAISS vector store as an MCP server – Agent Memory, drop-in local semantic search for Claude / Copilot / Agents.

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

Features

  • Local FAISS vector store as an MCP server – Agent Memory, dr

Use Cases

Create drop-in local semantic search for Claude and other AI agents.
Store agent memory in a local FAISS vector database.
Enable RAG workflows without cloud dependencies.
nonatofabio

Maintainer

LicenseMIT
Languagepython
Versionv1.0.0
UpdatedMay 8, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux
View SourceBrowse All Servers

Installation

Manual Installation

npx local-faiss-mcp

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 Local Faiss

The Local FAISS MCP server turns a local FAISS vector database into an MCP-accessible memory store for AI agents and assistants, enabling fully offline semantic search and retrieval-augmented generation (RAG) without any cloud dependencies. It exposes two core tools — ingest_document for indexing files (PDF, TXT, Markdown, and more with pandoc) and query_rag_store for semantic search — along with built-in prompts for cited answer extraction and document summarization. It is ideal for developers and researchers who want persistent agent memory or private document search running entirely on their own machine.

Prerequisites

  • Python 3.8+ and pip installed
  • pandoc installed via your system package manager (optional, for DOCX/HTML/EPUB support)
  • An MCP-compatible client such as Claude Desktop or Claude Code
1

Install the package

Install local-faiss-mcp from PyPI using pip.

pip install local-faiss-mcp
2

Install pandoc for extended format support (optional)

For indexing DOCX, HTML, or EPUB files, install pandoc using your system package manager. PDF and plain text work without pandoc.

# macOS
brew install pandoc

# Ubuntu/Debian
sudo apt install pandoc

# Windows
winget install pandoc
3

Add to Claude Code configuration

Create or edit .mcp.json in your project root to register the server. The --index-dir argument controls where the FAISS index and metadata are stored.

{
  "mcpServers": {
    "local-faiss-mcp": {
      "command": "local-faiss-mcp",
      "args": ["--index-dir", "./.vector_store"]
    }
  }
}
4

Add to Claude Desktop configuration (alternative)

For Claude Desktop, add the server block to ~/Library/Application Support/Claude/claude_desktop_config.json.

{
  "mcpServers": {
    "local-faiss-mcp": {
      "command": "local-faiss-mcp",
      "args": ["--index-dir", "/Users/yourname/vector_store", "--embed", "all-MiniLM-L6-v2"]
    }
  }
}
5

Ingest your first document and run a query

Use Claude to ingest a document into the vector store and then query it semantically.

Local Faiss Examples

Client configuration

Claude Code .mcp.json configuration for local-faiss-mcp with a custom vector store directory.

{
  "mcpServers": {
    "local-faiss-mcp": {
      "command": "local-faiss-mcp",
      "args": [
        "--index-dir", "./.vector_store",
        "--embed", "all-MiniLM-L6-v2"
      ]
    }
  }
}

Prompts to try

Example tasks to run with Claude once the local FAISS server is connected.

- "Ingest the file /Users/me/docs/research-paper.pdf into the vector store"
- "Search the vector store for information about transformer attention mechanisms and return the top 5 results"
- "What does the ingested documentation say about authentication? Extract a cited answer."
- "Summarize all documents in the vector store related to deployment configuration"
- "Ingest this text as a memory: 'The project launch date is March 15, 2026'"

Troubleshooting Local Faiss

ingest_document fails for DOCX or HTML files

These formats require pandoc to be installed and accessible in your PATH. Install pandoc via 'brew install pandoc' (macOS), 'apt install pandoc' (Linux), or from pandoc.org. PDF and plain text work without pandoc.

Embedding model download is slow on first run

The default model (all-MiniLM-L6-v2) is downloaded from Hugging Face on first use. This is a one-time download (~90 MB). Subsequent runs use the cached model. Ensure you have internet access during the first startup.

query_rag_store returns irrelevant results

Try lowering the top_k parameter to return fewer but more focused results. Also ensure documents were fully ingested — check for errors during the ingest_document step. For better semantic matches, rephrase queries to use terminology that appears in the source documents.

Frequently Asked Questions about Local Faiss

What is Local Faiss?

Local Faiss is a Model Context Protocol (MCP) server that local faiss vector store as an mcp server – agent memory, drop-in local semantic search for claude / copilot / agents. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Local Faiss?

Follow the installation instructions on the Local Faiss GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Local Faiss?

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

Is Local Faiss free to use?

Yes, Local Faiss is open source and available under the MIT 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

Local Faiss Alternatives — Similar Knowledge & Memory Servers

Looking for alternatives to Local Faiss? 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": { "local-faiss-mcp": { "command": "npx", "args": ["-y", "local-faiss-mcp"] } } }

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

Read the full setup guide →

Ready to use Local Faiss?

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