Chimera

You run Chimera MCP Server to expose a memory-focused AI backend that coordinates Notion data, builds a semantic graph in Neo4j, and serves clients via the MCP protocol for multi-client access. It enables fast, intelligent search, synchronized data, and secure, scalable querying for AI clients on your team.

How to use

You connect an MCP client to the Chimera server to perform intelligent Notion-based queries, share context across multiple clients, and leverage the server’s memory graph. Start the server, point clients at the MCP endpoint, and use the same API key across clients for authenticated access. The MCP server handles incrementally syncing data, building a semantic graph, and answering natural language queries through your AI clients.

How to install

Prerequisites you need before installation:

• Python 3.11+
• Neo4j 5.0+ database
• UV package manager
• Notion API Token
• Google Gemini API Key

1. Clone the Chimera repository and open the project folder.

# 1. Clone the repository
git clone <repo-url>
cd Chimera

# 2. Install UV (if not present)
curl -LsSf https://astral.sh/uv/install.sh | sh

# 3. Install dependencies with UV (recommended)
uv sync

# 4. Or install with pip
pip install -r requirements.txt

2. Create the environment configuration and fill in your values.

# Copy the example environment file
cp .env.example .env

# Then edit .env with your credentials
# Neo4j
NEO4J_URI=neo4j://127.0.0.1:7687
NEO4J_USERNAME=neo4j
NEO4J_PASSWORD=your_password

# Notion API
NOTION_TOKEN=your_notion_integration_token

# Google Gemini
GEMINI_API_KEY=your_gemini_api_key

# MCP server
CHIMERA_API_KEY=your_api_key

3. Start a Neo4j database (recommended via Docker) and wait for it to be ready.

# Docker-based Neo4j
docker run -d \
  --name neo4j \
  -p 7474:7474 -p 7687:7687 \
  -e NEO4J_AUTH=neo4j/your_password \
  neo4j:5.15

4. Start the Chimera system with the full run command.

uv run python run_chimera.py

Additional configurations and integration

Claude Desktop integration can be configured to connect to the MCP server. Add the following to claude_desktop_config.json to enable remote MCP access.

{
  "mcpServers": {
    "chimera-memory": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "http://localhost:3000/mcp/",
        "--header",
        "Authorization:Bearer your_api_key",
        "--allow-http"
      ]
    }
  }
}

Production deployment

Run the MCP server and the sync service in the background for a production setup.

ohup uv run python fastmcp_server.py --host 0.0.0.0 --port 3000 > logs/mcp_server.log 2>&1 &
ohup uv run python sync_service/sync_service.py > logs/sync_service.log 2>&1 &

# Check status
ps aux | grep fastmcp_server
ps aux | grep sync_service

# Stop services
pkill -f fastmcp_server
pkill -f sync_service

Environment variables and MCP API key

The server uses a central API key to authorize MCP clients. Ensure you provide the key in a secure environment variable and reference it in client configurations.