MCP Server for Mem.ai

A production-ready Model Context Protocol (MCP) server that provides AI assistants with intelligent access to Mem.ai's knowledge management platform.

Features

• Intelligent Memory: Save and process content with Mem It's AI-powered organization
• Note Management: Create, read, and delete structured markdown notes
• Collections: Organize notes into searchable collections
• Type-Safe: Full type hints and Pydantic validation
• Async/Await: High-performance async I/O throughout
• Clean API: Simple, intuitive interface for AI assistants
• Production-Ready: Comprehensive error handling and logging
• Well-Tested: Full test suite with pytest

Prerequisites

• Python 3.10 or higher
• A Mem.ai account
• Mem.ai API key (get one here)

Quick Start

Installation

1. Clone the repository:

git clone https://github.com/yourusername/mcp-mem.ai.git
cd mcp-mem.ai

1. Install dependencies:

pip install -e .

1. Set up your environment:

cp .env.example .env
# Edit .env and add your MEM_API_KEY

Running the Server

Local Development

fastmcp run src/mcp_mem/server.py

Using with Claude Desktop

Add to your Claude Desktop configuration (claude_desktop_config.json):

{
  "mcpServers": {
    "mem": {
      "command": "python",
      "args": ["-m", "mcp_mem.server"],
      "env": {
        "MEM_API_KEY": "your_api_key_here"
      }
    }
  }
}

Using with Other MCP Clients

from mcp_mem import mcp

# Run the server
mcp.run()

Available Tools

1. mem_it - Intelligent Content Processing

Save and automatically process any content type with AI-powered organization.

Parameters:

• input (required): Content to save (text, HTML, markdown, etc.)
• instructions (optional): Processing instructions
• context (optional): Additional context for organization
• timestamp (optional): ISO 8601 timestamp

Example:

mem_it(
    input="Just had a great meeting with the product team about Q1 roadmap...",
    instructions="Extract key action items and decisions",
    context="Product Planning"
)

2. create_note - Create Structured Note

Create a markdown-formatted note with explicit control over content and organization.

Parameters:

• content (required): Markdown-formatted content
• collection_ids (optional): List of collection UUIDs
• collection_titles (optional): List of collection titles

Example:

create_note(
    content="""# Team Standup - Jan 15, 2024

    ## Completed
    - Feature X shipped to production
    - Bug fixes for issue #123

    ## In Progress
    - Working on Feature Y
    - Code review for PR #456

    ## Blockers
    - Waiting for API access
    """,
    collection_titles=["Team Meetings", "Engineering"]
)

3. read_note - Read Note

Retrieve a note's full content and metadata by ID.

Parameters:

• note_id (required): UUID of the note

Example:

read_note("01961d40-7a67-7049-a8a6-d5638cbaaeb9")

4. delete_note - Delete Note

Permanently delete a note by ID.

Parameters:

• note_id (required): UUID of the note

Example:

delete_note("01961d40-7a67-7049-a8a6-d5638cbaaeb9")

5. create_collection - Create Collection

Create a new collection to organize related notes.

Parameters:

• title (required): Collection title
• description (optional): Markdown-formatted description

Example:

create_collection(
    title="Project Apollo",
    description="""# Project Apollo

    All notes related to the Apollo project including:
    - Meeting notes
    - Technical specifications
    - Customer feedback
    """
)

6. delete_collection - Delete Collection

Delete a collection (notes remain, just unassociated).

Parameters:

• collection_id (required): UUID of the collection

Example:

delete_collection("5e29c8a2-c73b-476b-9311-e2579712d4b1")

Configuration

Configuration is done via environment variables. Copy .env.example to .env and customize:

# Required: Your Mem.ai API key
MEM_API_KEY=your_api_key_here

# Optional: Custom API endpoint (default: https://api.mem.ai/v2)
MEM_API_BASE_URL=https://api.mem.ai/v2

# Optional: Request timeout in seconds (default: 30)
MEM_REQUEST_TIMEOUT=30

# Optional: Enable debug logging (default: false)
MEM_DEBUG=false

Architecture

src/mcp_mem/
 __init__.py      # Package initialization
 models.py        # Pydantic data models
 client.py        # Mem.ai API client
 server.py        # MCP server implementation

Key Components

• models.py: Pydantic models for request/response validation
• client.py: Async HTTP client wrapper for Mem.ai API
• server.py: FastMCP server with tool implementations

Testing

Run the test suite:

# Install dev dependencies
pip install -e ".[dev]"

# Run all tests
pytest

# Run with coverage
pytest --cov=mcp_mem --cov-report=html

# Run specific test file
pytest tests/test_client.py

Error Handling

The server provides clear, actionable error messages:

• MemAuthenticationError: Invalid or missing API key
• MemNotFoundError: Resource (note/collection) not found
• MemValidationError: Invalid request parameters
• MemAPIError: General API errors

All errors are logged and returned with helpful context to the AI assistant.

Examples

See the examples/ directory for complete usage examples:

• basic_usage.py: Simple examples of each tool
• advanced_usage.py: Complex workflows and patterns

Contributing

Contributions are welcome! Please feel free to submit a …