MemPalace
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
What is MemPalace?
MemPalace is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to 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 ...
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
This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- A local AI memory system that stores all conversations verba
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx mempalaceConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use MemPalace
MemPalace is a fully local AI memory system that stores conversation history verbatim using ChromaDB and organizes memories into a navigable hierarchy of Wings (people/projects), Rooms (topics), and Halls (memory types like decisions and discoveries). It achieves a 96.6% recall score on LongMemEval benchmarks with no API calls, cloud services, or subscriptions required. Through its 19 MCP tools, AI assistants like Claude can search past decisions, retrieve debugging sessions, query a temporal knowledge graph, and write diary entries — giving any LLM persistent, structured memory across sessions.
Prerequisites
- Python 3.8+ installed on your system
- pip package manager
- No API keys or cloud accounts needed — everything runs locally
- An MCP client such as Claude Desktop or Claude Code
- Sufficient disk space for your conversation history (ChromaDB stores data locally)
Install MemPalace
Install the mempalace package from PyPI. It bundles ChromaDB and all dependencies — no separate database server is required.
pip install mempalaceInitialize a palace for your project
Create a new palace (local memory store) for a specific project or directory. This sets up the ChromaDB collection and wing/room structure.
mempalace init ~/projects/myappMine existing conversations into memory
Import your existing chat exports (Claude, ChatGPT, Slack logs) or code/docs into the palace. Use --mode convos for chat logs and --extract general to auto-classify into decisions, milestones, and problems.
mempalace mine ~/chats/ --mode convos --extract general
mempalace mine ~/projects/myapp # For code and docsRegister the MCP server with your AI client
Add MemPalace to your MCP client using the claude mcp add command or by editing the config file manually. The server runs as a Python module.
claude mcp add mempalace -- python -m mempalace.mcp_serverLoad your memory context at the start of a session
Run wake-up to generate a compact summary of critical facts (~170 tokens) that you can paste into your AI assistant's context, or rely on the MCP tools for on-demand retrieval.
mempalace wake-up
mempalace statusMemPalace Examples
Client configuration
Add MemPalace to claude_desktop_config.json. The server runs locally as a Python module with no network dependencies.
{
"mcpServers": {
"mempalace": {
"command": "python",
"args": ["-m", "mempalace.mcp_server"],
"env": {
"MEMPALACE_PATH": "/Users/yourname/.mempalace"
}
}
}
}Prompts to try
Use these prompts after connecting MemPalace to have your AI assistant search and retrieve past context from your memory palace.
- "Search my memory palace for why we switched to GraphQL instead of REST"
- "What decisions did we make about the authentication system in the myapp project?"
- "Show me the knowledge graph timeline for the Driftwood project"
- "File this conversation to my memory palace under the 'myapp' wing, room 'architecture'"
- "What problems did we encounter with database migrations last quarter?"Troubleshooting MemPalace
mempalace search returns no results after mining conversations
Verify the palace was initialized at the correct path and that mining completed without errors. Run 'mempalace status' to confirm the wing/room structure exists and contains drawers. Check that MEMPALACE_PATH points to the same directory used during init.
MCP tools are not visible in Claude Desktop
Ensure the Python environment with mempalace installed is the one being invoked. Use the full path to python if needed. Check 'python -m mempalace.mcp_server --help' in a terminal to confirm the module loads correctly.
ChromaDB errors on startup about locked collections
ChromaDB uses file locks and only supports one writer at a time. Close any other processes using the palace (e.g. a running mempalace search CLI command) before starting the MCP server.
Frequently Asked Questions about MemPalace
What is MemPalace?
MemPalace is a Model Context Protocol (MCP) server that 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 It connects AI assistants to external tools and data sources through a standardized interface.
How do I install MemPalace?
Follow the installation instructions on the MemPalace GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with MemPalace?
MemPalace works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is MemPalace free to use?
Yes, MemPalace is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.
MemPalace Alternatives — Similar Knowledge & Memory Servers
Looking for alternatives to MemPalace? Here are other popular knowledge & memory servers you can use with Claude, Cursor, and VS Code.
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
★ 15.4kAn 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
★ 13.7kMemory for 24/7 proactive agents like OpenClaw.
MemOS
★ 9.3kMemOS (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
★ 5.4kBuild, evaluate, and integrate long-term memory for self-evolving agents.
Cipher
★ 4.8kCipher is an opensource memory layer specifically designed for coding agents. Compatible with Cursor, Windsurf, Claude Desktop, Claude Code, Gemini CLI, AWS's Kiro, VS Code, and Roo Code through MCP, and coding agents, such as Kimi K2.
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.
Set Up MemPalace in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use MemPalace?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.