filemaker-odata-mcp
MCP server for FileMaker Server OData 4.01 API integration, enabling AI assistants to discover databases, perform CRUD operations, and manage connections.
README
FileMaker Server OData MCP
Model Context Protocol (MCP) server providing FileMaker Server OData 4.01 API integration for AI assistants like Claude Desktop, Windsurf, Cursor, and Cline.
Features
- 🔌 19 MCP Tools for FileMaker database operations
- 🔍 Database Discovery - Explore tables, fields, and metadata
- 📊 CRUD Operations - Create, read, update, and delete records
- 🔐 Secure Connections - SSL support for self-signed certificates
- 💾 Connection Management - Save and reuse database connections
- 📝 OData 4.01 Standard - Full query capabilities ($filter, $select, $orderby, etc.)
Quick Start
Installation
# Via NPM (recommended)
npm install -g filemaker-odata-mcp
# Or local development
git clone https://github.com/fsans/FMS-ODATA-MCP.git
cd FMS-ODATA-MCP
npm install
npm run build
Deployment Modes
1. MCP Server Mode (Default)
For use with AI assistants that support MCP (Claude Desktop, Windsurf, Cursor, Cline).
Setup for Claude Desktop
-
Locate your Claude config file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
- macOS:
-
Add the MCP server:
{
"mcpServers": {
"filemaker": {
"command": "npx",
"args": ["-y", "filemaker-odata-mcp"],
"env": {
"FM_SERVER": "https://your-filemaker-server.com",
"FM_DATABASE": "YourDatabase",
"FM_USER": "your-username",
"FM_PASSWORD": "your-password",
"FM_VERIFY_SSL": "true"
}
}
}
}
-
For self-signed SSL certificates, set
FM_VERIFY_SSLto"false" -
Restart Claude Desktop
Setup for Windsurf/Cursor
The server will be automatically detected when installed globally. For local development, add to your MCP config.
2. Standalone HTTP Server Mode
Run as a standalone HTTP server accessible from any application:
# Set environment variables for HTTP mode
export MCP_TRANSPORT=http
export MCP_PORT=3333
export MCP_HOST=0.0.0.0 # Listen on all interfaces
# Run the server
filemaker-odata-mcp
The server will start on http://localhost:3333 with the following endpoints:
- MCP endpoint:
http://localhost:3333/mcp(POST requests with JSON-RPC 2.0) - Health check:
http://localhost:3333/health - Server info:
http://localhost:3333/mcp(GET request)
Example HTTP Client Request
curl -X POST http://localhost:3333/mcp \
-H "Content-Type: application/json" \
-d '{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/list",
"params": {}
}'
HTTPS Mode
export MCP_TRANSPORT=https
export MCP_PORT=3443
export MCP_CERT_PATH=/path/to/cert.pem
export MCP_KEY_PATH=/path/to/key.pem
Integration Examples
Python Example:
import requests
# List available tools
response = requests.post("http://localhost:3333/mcp", json={
"jsonrpc": "2.0",
"id": 1,
"method": "tools/list",
"params": {}
})
tools = response.json()
# Query records
response = requests.post("http://localhost:3333/mcp", json={
"jsonrpc": "2.0",
"id": 2,
"method": "tools/call",
"params": {
"name": "fm_odata_query_records",
"arguments": {
"table": "Contacts",
"filter": "City eq 'New York'"
}
}
})
JavaScript Example:
// Connect to FileMaker
const response = await fetch('http://localhost:3333/mcp', {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({
jsonrpc: '2.0',
id: 1,
method: 'tools/call',
params: {
name: 'fm_odata_connect',
arguments: {
server: 'https://your-server.com/fmi/odata/v4',
database: 'Contacts',
user: 'admin',
password: 'secret',
verifySsl: false
}
}
})
});
3. Docker Deployment
Option A: Using start.sh (Recommended)
The included start.sh script handles building, credential injection from .env, and container lifecycle:
git clone https://github.com/fsans/FMS-ODATA-MCP.git
cd FMS-ODATA-MCP
cp .env.example .env
# Edit .env with your FileMaker credentials and set MCP_TRANSPORT=http, MCP_HOST=0.0.0.0
./start.sh
The script will build TypeScript, build the Docker image, remove any existing container, and start a fresh one. Logs are tailed automatically.
Option B: Using Docker Run
# Clone and build
git clone https://github.com/fsans/FMS-ODATA-MCP.git
cd FMS-ODATA-MCP
npm run build
docker build -t filemaker-odata-mcp:latest .
# Run the container
docker run -d \
--name filemaker-odata-mcp \
-p 3333:3333 \
-e FM_SERVER=https://your-filemaker-server.com \
-e FM_DATABASE=YourDatabase \
-e FM_USER=your-username \
-e FM_PASSWORD=your-password \
-e FM_VERIFY_SSL=false \
-e MCP_TRANSPORT=http \
-e MCP_HOST=0.0.0.0 \
-v ~/.fms-odata-mcp:/home/mcp/.fms-odata-mcp \
filemaker-odata-mcp:latest
Important: Set
MCP_HOST=0.0.0.0when running in a container. Usinglocalhostbinds only to the container's loopback interface and makes the port unreachable from outside.
Option C: Using Docker Compose
- Clone and build:
git clone https://github.com/fsans/FMS-ODATA-MCP.git
cd FMS-ODATA-MCP
npm run build
- Configure environment:
cp docker-compose.yml my-docker-compose.yml
# Edit my-docker-compose.yml with your FileMaker credentials
- Start the server:
docker-compose -f my-docker-compose.yml up -d
Option D: Docker Compose with HTTPS
# Start with Nginx reverse proxy for HTTPS
docker-compose -f docker-compose.yml --profile https up -d
Place your SSL certificates in the ./ssl directory:
ssl/cert.pem- SSL certificatessl/key.pem- SSL private key
Docker Features
- Health checks - Automatic monitoring of server status
- Persistent connections - Mount volume to save connection configurations
- Non-root user - Security best practices
- Alpine Linux - Small image size (~50MB)
- Signal handling - Graceful shutdown with dumb-init
Accessing the Server
Once running, access the server at:
- HTTP:
http://localhost:3333 - HTTPS (with Nginx):
https://localhost
Check the health endpoint:
curl http://localhost:3333/health
4. Dify Integration
To use this server as an MCP tool in Dify:
- Start the server in HTTP mode (see Docker Deployment above)
- In Dify, add a new MCP tool with:
- Transport:
streamable_http - URL:
http://host.docker.internal:3333/mcp(if Dify runs in Docker on the same host)
- Transport:
- No authentication headers are required
Note: If Dify returns a 403 error, check your SSRF proxy configuration. Dify uses a Squid proxy to prevent SSRF attacks — port 3333 must be added to the
Safe_portsACL insquid.conf.
First Steps
Once connected, try these prompts in Claude:
What tables are in my FileMaker database?
Show me the first 5 records from the Contacts table
Find all contacts where LastName equals "Smith"
Create a new contact with name "John Doe" and email "john@example.com"
Documentation
- Quick Reference - One-page setup guide
- Prompt Examples - Complete prompt reference
- Claude Desktop Setup - Detailed configuration
- Windsurf Setup - IDE integration guide
- Docker Deployment - Complete Docker guide with production examples
Available Tools
| Category | Tools |
|---|---|
| Discovery | fm_odata_list_tables, fm_odata_get_metadata, fm_odata_get_service_document |
| Queries | fm_odata_query_records, fm_odata_get_record, fm_odata_get_records, fm_odata_count_records |
| CRUD | fm_odata_create_record, fm_odata_update_record, fm_odata_delete_record |
| Connection | fm_odata_connect, fm_odata_set_connection, fm_odata_list_connections, fm_odata_get_current_connection |
| Config | fm_odata_config_add_connection, fm_odata_config_remove_connection, fm_odata_config_list_connections |
Requirements
- Node.js 18.0.0 or higher
- FileMaker Server with OData API enabled
- FileMaker Account with appropriate access privileges
Environment Variables
FileMaker Connection
| Variable | Description | Required | Default |
|---|---|---|---|
FM_SERVER |
FileMaker Server URL | Yes | - |
FM_DATABASE |
Database name | Yes | - |
FM_USER |
Username | Yes | - |
FM_PASSWORD |
Password | Yes | - |
FM_VERIFY_SSL |
Verify SSL certificates | No | true |
FM_TIMEOUT |
Request timeout (ms) | No | 30000 |
HTTP/HTTPS Transport
| Variable | Description | Required | Default |
|---|---|---|---|
MCP_TRANSPORT |
Transport type: stdio, http, or https |
No | stdio |
MCP_PORT |
Port for HTTP/HTTPS server | No | 3333 (HTTP), 3443 (HTTPS) |
MCP_HOST |
Host to bind to | No | localhost |
MCP_CERT_PATH |
Path to SSL certificate (HTTPS only) | No | - |
MCP_KEY_PATH |
Path to SSL private key (HTTPS only) | No | - |
OData Query Syntax
The server supports OData 4.01 query options:
$filter - Filter records (e.g., "Age gt 18")
$select - Select specific fields
$orderby - Sort results
$top - Limit results
$skip - Skip records (pagination)
$expand - Include related records
$count - Include total count
Example prompts:
Get contacts where Age is greater than 18
Show only Name and Email fields from Contacts
Sort contacts by LastName in descending order
Get the first 10 contacts, skip the first 20
Contributing
See CONTRIBUTING.md for development guidelines.
License
MIT License - see LICENSE file for details.
Support
- Issues: GitHub Issues
- Discussions: GitHub Discussions
Changelog
See dev_stuf/VERSIONING.md for version history.
Made with ❤️ for the FileMaker and AI communities
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.