DocketBird MCP Server

This MCP server provides access to DocketBird's court case data and document management functionality.

Requirements

Python 3.11
uv package manager

Setup

Install uv if you haven't already:

curl -LsSf https://astral.sh/uv/install.sh | sh

Create and activate a virtual environment:

uv venv
source .venv/bin/activate  # On Unix/MacOS
# OR
.venv\Scripts\activate     # On Windows

Install dependencies:

uv pip install .

Set up your environment variables:

export DOCKETBIRD_API_KEY=your_api_key_here  # On Unix/MacOS
# OR
set DOCKETBIRD_API_KEY=your_api_key_here     # On Windows

Running the Server

Run the server using:

uv run docketbird_mcp.py --transport stdio  # For stdio transport
uv run docketbird_mcp.py --transport sse    # For SSE transport

Available Tools

The server provides the following tools:

get_case_details: Get comprehensive details about a case including all documents
download_document_by_id: Download a specific document by its DocketBird ID
list_cases: Get a list of cases belonging to an account
list_courts_and_types: Get a comprehensive list of all available courts and case types

Configuration Files

Make sure these files are in the same directory as the script:

courts.json: Contains information about all available courts
case_types.json: Contains information about different types of cases

MCP Server Configuration

The MCP server configuration can be added to one of these locations depending on your MCP client:

Cursor: ~/.cursor/mcp.json
Claude in mac: ~/Library/Application Support/Claude/claude_desktop_config.json

Install uv if you haven't already:

curl -LsSf https://astral.sh/uv/install.sh | sh

Add the following configuration to the appropriate file:

{
  "mcpServers": {
    "docketbird-mcp": {
            "command": "uv",
            "args": [
                "run",
                "--directory",
                "PATH_TO_THE_SERVER/docketbird-mcp",
                "python",
                "docketbird_mcp.py"],
            "env": {
                "DOCKETBIRD_API_KEY": "YOUR_KEY"
            }
        }
}

Deployment

The DocketBird MCP server can be deployed to a cloud server using Docker and GitHub Actions. The deployment process is defined in the .github/workflows/deploy.yml file.

Docker Deployment

The server is containerized using Docker. You can build and run the Docker image locally with the desired transport type:

# Build for ARM architecture (M1/M2 Mac)
docker buildx build --platform linux/arm64 -t docketbird-mcp-arm:latest --load .

# Build for AMD architecture (standard servers)
docker buildx build --platform linux/amd64 -t docketbird-mcp:latest --load .

# Run locally with stdio transport
docker run -d \
  --name docketbird-mcp-stdio \
  --restart=always \
  -e DOCKETBIRD_API_KEY="your_api_key_here" \
  -e TRANSPORT_TYPE="stdio" \
  docketbird-mcp-arm:latest /app/start.sh

# Run locally with SSE transport
docker run -d \
  --name docketbird-mcp-sse \
  --restart=always \
  -e DOCKETBIRD_API_KEY="your_api_key_here" \
  -e TRANSPORT_TYPE="sse" \
  docketbird-mcp-arm:latest /app/start.sh

Validating Deployment

To validate that your deployment is working correctly:

Check that the container is running:

docker ps | grep docketbird-mcp

Verify the container logs:

docker logs docketbird-mcp

The logs should show:

Starting DocketBird MCP server...
API Key set: your_...
Running python docketbird_mcp.py

Test the connection from your MCP client using the configuration from this README.

If the container isn't running, you can troubleshoot by checking:

Docker image exists: docker images | grep docketbird
Container logs for errors: docker logs docketbird-mcp
Server logs: Check if there are any permission or network issues

DocketBird MCP Server