MemDocs

Provides a local, git-native memory server that serves persistent project memory via MCP for real-time AI context.

python

1

GitHub Stars

python

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "smart-ai-memory-memdocs": {
      "command": "memdocs",
      "args": [
        "serve",
        "--mcp"
      ]
    }
  }
}

MemDocs offers a persistent, git-native memory layer that serves project context to AI assistants using the Model Context Protocol (MCP). It enables real-time memory access, cross-session learning, and offline operation, making your AI copilots smarter and more proactive without relying on cloud memory.

How to use

Set up MemDocs to run an MCP server locally, so your AI tools can request the project memory you store in your repository. You will start a memory context server that serves memory on demand and can auto-start when your project is opened in supporting IDEs.

How to install

Prerequisites include Python and a working Git environment. You will install MemDocs from PyPI, optionally with embeddings for local vector search.

# From PyPI (recommended)
pip install memdocs

# With optional features
pip install memdocs[embeddings]  # Local vector search
pip install memdocs[all]         # All features

# From source (optional for development)
git clone https://github.com/Smart-AI-Memory/memdocs.git
cd memdocs
pip install -e ".[dev,embeddings]"

MCP Server setup and usage

To enable the MCP server, start MemDocs with the MCP mode. This serves memory to AI assistants in real time and can be integrated with Claude Desktop or other MCP-compatible clients.

# Start MCP server
memdocs serve --mcp

# Or auto-start in VS Code by adding a suitable task
# See guides for details on configuring your workspace

Basic workflow with MCP

After starting the MCP server, you can access and share project memory with compatible clients. Use MemDocs commands to initialize memory, review changes, and query for context to feed into your AI sessions.

# Initialize MemDocs in your project (MCP enabled by default)
mcdocs init

# Start memory serving for MCP
memdocs serve --mcp

# Open your project in an MCP-enabled environment and interact with memory
# Use query, review, and stats to manage context

Available tools

init

Initializes MemDocs in the current project, preparing the memory store and MCP integration.

review

Generates memory documentation for changed files or specified paths to keep the project memory up to date.

query

Searches the local memory store for relevant context to feed AI sessions.

stats

Shows memory usage and statistics for the MemDocs store.

export

Exports memory representations to other formats for integration with external tools.

serve

Runs the MCP server, exposing memory to MCP-enabled AI clients.

help

Displays help and usage information for MemDocs commands.