MemoryMesh

v1.0.0Knowledge & Memorystable

A universal, local-first MCP hub that indexes personal files (documents, code, etc.) and provides private semantic search via hybrid dense+BM25 retrieval, enabling agents like Claude Desktop to query your data without sending it to the cloud.

aimcp
Share:
342
Stars
0
Downloads
0
Weekly
0/5

What is MemoryMesh?

MemoryMesh is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to universal, local-first mcp hub that indexes personal files (documents, code, etc.) and provides private semantic search via hybrid dense+bm25 retrieval, enabling agents like claude desktop to query yo...

A universal, local-first MCP hub that indexes personal files (documents, code, etc.) and provides private semantic search via hybrid dense+BM25 retrieval, enabling agents like Claude Desktop to query your data without sending it to the cloud.

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

Features

  • A universal, local-first MCP hub that indexes personal files

Use Cases

Local-first semantic search
Personal file indexing
Private AI knowledge base
CheMiguel23

Maintainer

LicenseMIT License
Languagetypescript
Versionv1.0.0
UpdatedMay 13, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y memorymesh

Manual Installation

npx -y memorymesh

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 MemoryMesh

MemoryMesh is a local-first MCP server that provides a structured knowledge graph for AI agents, using schema-driven node and edge definitions to store and retrieve information entirely on your machine. It automatically generates tools from JSON schema files, allowing Claude Desktop and other MCP clients to create, update, delete, and query typed entities and relationships without sending data to external services. It ships with 11 pre-built schemas designed for text-based RPG game state management, and can be extended with custom schemas for any domain where persistent, structured memory across AI sessions is needed.

Prerequisites

  • Node.js 18+ installed
  • Git installed for cloning the repository
  • An MCP-compatible client such as Claude Desktop
  • Basic understanding of JSON schema format (for custom schemas)
1

Clone the MemoryMesh repository

Clone the repository to your local machine. MemoryMesh runs from source as a Node.js process.

git clone https://github.com/CheMiguel23/memorymesh.git && cd memorymesh
2

Install dependencies

Install the required Node.js packages.

npm install
3

Build the TypeScript project

Compile the TypeScript source to JavaScript. The output goes to the `dist/` directory.

npm run build
4

Configure Claude Desktop

Add MemoryMesh to your Claude Desktop configuration file, using the absolute path to the compiled entry point. On macOS this file is at `~/Library/Application Support/Claude/claude_desktop_config.json`.

{
  "mcpServers": {
    "memorymesh": {
      "command": "node",
      "args": ["/absolute/path/to/memorymesh/dist/index.js"]
    }
  }
}
5

Restart Claude Desktop

Close and reopen Claude Desktop to load the MemoryMesh server. The 11 pre-built RPG schemas are active immediately — you will see tools like `add_npc`, `update_location`, `delete_artifact` available.

6

Create custom schemas (optional)

To use MemoryMesh for domains other than RPG game state, create `.schema.json` files in `dist/data/schemas/`. Each schema definition automatically generates `add_`, `update_`, and `delete_` tools for that entity type.

// Example: dist/data/schemas/task.schema.json
{
  "name": "task",
  "fields": [
    { "name": "title", "type": "string", "required": true },
    { "name": "status", "type": "string" },
    { "name": "priority", "type": "string" }
  ]
}

MemoryMesh Examples

Client configuration

Claude Desktop configuration for MemoryMesh. Replace the path with the absolute path to your local clone's dist/index.js file.

{
  "mcpServers": {
    "memorymesh": {
      "command": "node",
      "args": ["/Users/you/memorymesh/dist/index.js"]
    }
  }
}

Prompts to try

These prompts show MemoryMesh in action for an RPG campaign, but the same patterns apply to any custom schema you define:

- "Add an NPC named 'Elder Mira' who is a wise herbalist living in the village of Thornhaven"
- "Update the location 'Ancient Forest' to note that the players discovered a hidden shrine there"
- "Show me all NPCs in the current session and their relationships to the main quest"
- "Create a relationship between Elder Mira and the Ancient Forest — she gathers herbs there weekly"
- "Delete the artifact 'Cursed Amulet' from the knowledge graph — it was destroyed in the last session"

Troubleshooting MemoryMesh

Claude Desktop shows no MemoryMesh tools after restart

Check that the path in `claude_desktop_config.json` is an absolute path (not relative), points to `dist/index.js` (not `src/`), and that `npm run build` completed without errors. On macOS, the config file location is `~/Library/Application Support/Claude/claude_desktop_config.json`.

Schema tools appear but data is lost between Claude sessions

MemoryMesh stores data in `dist/data/memory.json`. Verify this file exists and is writable. You can customize the storage path by editing `MEMORY_FILE` in `/config/config.ts` and rebuilding. Do not delete the dist/ directory between sessions.

Custom schema tools are not appearing after adding a new schema file

Schema files must be placed in `dist/data/schemas/` (the compiled output directory), not in `src/`. Also ensure the file name ends with `.schema.json` and that the JSON is valid. Restart Claude Desktop after adding new schemas.

Frequently Asked Questions about MemoryMesh

What is MemoryMesh?

MemoryMesh is a Model Context Protocol (MCP) server that universal, local-first mcp hub that indexes personal files (documents, code, etc.) and provides private semantic search via hybrid dense+bm25 retrieval, enabling agents like claude desktop to query your data without sending it to the cloud. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install MemoryMesh?

Install via npm with the command: npx -y memorymesh. 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 MemoryMesh?

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

Is MemoryMesh free to use?

Yes, MemoryMesh is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

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.

Quick Config Preview

{ "mcpServers": { "memorymesh": { "command": "npx", "args": ["-y", "memorymesh"] } } }

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

Read the full setup guide →

Ready to use MemoryMesh?

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