Synapse MCP Server

Synapse MCP Server

A Model Context Protocol server that exposes Synapse Entities (Datasets, Projects, Folders, Files, Tables) with their annotations, enabling programmatic access to Synapse data resources through a RESTful API.

Category
Visit Server

README

Synapse MCP Server

A Model Context Protocol (MCP) server that exposes Synapse Entities (Datasets, Projects, Folders, Files, Tables) with their annotations and supports OAuth2 authentication.

Overview

This server provides a RESTful API for accessing Synapse entities and their annotations through the Model Context Protocol (MCP). It allows you to:

  • Authenticate with Synapse
  • Retrieve entities by ID
  • Retrieve entities by name
  • Get entity annotations
  • Get entity children
  • Query entities based on various criteria
  • Query Synapse tables
  • Get datasets in Croissant metadata format

Installation

# Clone the repository
git clone https://github.com/SageBionetworks/synapse-mcp.git
cd synapse-mcp

# Create a virtual environment
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install dependencies
pip install -e .

Installing from PyPI

# Install from PyPI
pip install synapse-mcp

Usage

Starting the Server

python server.py --host 127.0.0.1 --port 9000

This will start the MCP server on the default port (9000).

Using the CLI

# Start the server using the CLI
synapse-mcp --host 127.0.0.1 --port 9000 --debug

Command-line Options

usage: server.py [-h] [--host HOST] [--port PORT] [--debug]

Run the Synapse MCP server with OAuth2 support

options:
  -h, --help       show this help message and exit
  --host HOST      Host to bind to
  --port PORT      Port to listen on
  --debug          Enable debug logging
  --server-url URL Public URL of the server (for OAuth2 redirect)

Running Tests

# Run all tests with coverage
./run_tests.sh

# Or run pytest directly
python -m pytest

Testing the Server

python examples/client_example.py

Authentication Methods

Environment Variables

The server supports the following environment variables:

  • HOST: The host to bind to (default: 127.0.0.1)
  • PORT: The port to listen on (default: 9000)
  • MCP_TRANSPORT: The transport protocol to use (default: stdio)
    • stdio: Use standard input/output for local development
    • sse: Use Server-Sent Events for cloud deployment
  • MCP_SERVER_URL: The public URL of the server (default: mcp://127.0.0.1:9000)
    • Used for OAuth2 redirect and server information

The server supports two authentication methods:

  1. Auth Token: Authenticate using a Synapse authentication token
  2. OAuth2: Authenticate using Synapse's OAuth2 server
    • Requires registering an OAuth2 client in Synapse (https://www.synapse.org/#!PersonalAccessTokens:OAuth)

API Endpoints

Server Information

  • GET /info - Get server information

Tools

  • GET /tools - List available tools
  • POST /tools/authenticate - Authenticate with Synapse
  • POST /tools/get_oauth_url - Get the OAuth2 authorization URL
  • POST /tools/get_entity - Get an entity by ID or name
  • POST /tools/get_entity_annotations - Get annotations for an entity
  • POST /tools/get_entity_children - Get child entities of a container entity
  • POST /tools/query_entities - Query entities based on various criteria
  • POST /tools/query_table - Query a Synapse table

Resources

  • GET /resources - List available resources
  • GET /resources/entity/{id} - Get entity by ID
  • GET /resources/entity/{id}/annotations - Get entity annotations
  • GET /resources/entity/{id}/children - Get entity children
  • GET /resources/query/entities/{entity_type} - Query entities by type
  • GET /resources/query/entities/parent/{parent_id} - Query entities by parent ID
  • GET /resources/query/entities/name/{name} - Query entities by name
  • GET /resources/query/table/{id}/{query} - Query a table with SQL-like syntax

OAuth2 Endpoints

  • GET /oauth/login - Redirect to Synapse OAuth2 login page
  • GET /oauth/callback - Handle OAuth2 callback from Synapse

Examples

Authentication

You need to authenticate with real Synapse credentials to use the server:

import requests

# Authenticate with Synapse
response = requests.post("http://127.0.0.1:9000/tools/authenticate", json={
    "email": "your-synapse-email@example.com",
    "password": "your-synapse-password"
})
result = response.json()
print(result)

# Alternatively, you can authenticate with an API key
response = requests.post("http://127.0.0.1:9000/tools/authenticate", json={
    "api_key": "your-synapse-api-key"
})

OAuth2 Authentication

1. Redirect Flow (Browser-based)

Direct users to the OAuth login URL:

http://127.0.0.1:9000/oauth/login?client_id=YOUR_CLIENT_ID&redirect_uri=YOUR_REDIRECT_URI

2. API-based Flow

For programmatic use, first get the authorization URL:

import requests

# Get OAuth2 authorization URL
response = requests.post("http://127.0.0.1:9000/tools/get_oauth_url", json={
    "client_id": "YOUR_CLIENT_ID",
    "redirect_uri": "YOUR_REDIRECT_URI"
})
auth_url = response.json()["auth_url"]
# Redirect user to auth_url

Getting an Entity

import requests

# Get an entity by ID
response = requests.get("http://127.0.0.1:9000/resources/entity/syn123456")  # Replace with a real Synapse ID
entity = response.json()
print(entity)

Getting Entity Annotations

import requests

# Get annotations for an entity
response = requests.get("http://127.0.0.1:9000/resources/entity/syn123456/annotations")  # Replace with a real Synapse ID
annotations = response.json()
print(annotations)

Querying Entities

import requests

# Query for files in a project
response = requests.get("http://127.0.0.1:9000/resources/query/entities/parent/syn123456", params={  # Replace with a real Synapse ID
    "entity_type": "file"
})
files = response.json()
print(files)

Querying a Table

import requests

# Query a table
table_id = "syn123456"  # Replace with a real Synapse table ID
query = "SELECT * FROM syn123456 LIMIT 10"  # Replace with a real Synapse table ID
response = requests.get(f"http://127.0.0.1:9000/resources/query/table/{table_id}/{query}")
table_data = response.json()
print(table_data)

Getting Datasets in Croissant Format

import requests
import json

# Get public datasets in Croissant format
response = requests.get("http://127.0.0.1:9000/resources/croissant/datasets")
croissant_data = response.json()

# Save to file
with open("croissant_metadata.json", "w") as f:
    json.dump(croissant_data, f, indent=2)

Deployment

Docker

You can build and run the server using Docker:

# Build the Docker image
docker build -t synapse-mcp .

# Run the container
docker run -p 9000:9000 -e SYNAPSE_OAUTH_CLIENT_ID=your_client_id -e SYNAPSE_OAUTH_CLIENT_SECRET=your_client_secret -e SYNAPSE_OAUTH_REDIRECT_URI=your_redirect_uri synapse-mcp
docker run -p 9000:9000 -e MCP_TRANSPORT=sse -e MCP_SERVER_URL=mcp://your-domain:9000 synapse-mcp

Fly.io

Deploy to fly.io:

# Install flyctl
curl -L https://fly.io/install.sh | sh

# Login to fly.io
flyctl auth login

# Launch the app
flyctl launch

# Set OAuth2 secrets
flyctl secrets set SYNAPSE_OAUTH_CLIENT_ID=your_client_id
flyctl secrets set SYNAPSE_OAUTH_CLIENT_SECRET=your_client_secret
flyctl secrets set SYNAPSE_OAUTH_REDIRECT_URI=https://your-app-name.fly.dev/oauth/callback
flyctl secrets set MCP_TRANSPORT=sse
flyctl secrets set MCP_SERVER_URL=mcp://your-app-name.fly.dev:9000

# Deploy
flyctl deploy

Integrating with Claude Desktop

You can integrate this Synapse MCP server with Claude Desktop to enable Claude to access and work with Synapse data directly in your conversations.

Setup Instructions

  1. First, clone the repository and install the requirements:
# Clone the repository
git clone https://github.com/susheel/synapse-mcp.git
cd synapse-mcp

# Create a virtual environment
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install dependencies
pip install -e .
  1. Configure Claude Desktop to use the Synapse MCP server:

    • Open Claude Desktop
    • Click on the Claude menu and select "Settings..."
    • Click on "Developer" in the left-hand bar
    • Click on "Edit Config"
    • Add the following configuration to the mcpServers section:
"synapse-mcp": {
  "command": "python",
  "args": [
    "/path/to/synapse-mcp/server.py",
    "--host", "127.0.0.1",
    "--port", "9000"
  ]
}
  1. Save the configuration file and restart Claude Desktop

  2. You can now use Synapse data in your conversations with Claude. For example:

    • "Get the entity with ID syn123456 from Synapse"
    • "Query all files in the Synapse project syn123456"
    • "Get annotations for the Synapse entity syn123456"

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT

Recommended Servers

playwright-mcp

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.

Official
Featured
TypeScript
Magic Component Platform (MCP)

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.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

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.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

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.

Official
Featured
TypeScript
Kagi MCP Server

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.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

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.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured