Skyvia MCP Server

A Model Context Protocol (MCP) server for interacting with the Skyvia API, focusing on integration-related endpoints.

Overview

This MCP server provides tools to interact with the Skyvia API, allowing you to:

• Manage workspaces
• Manage connections
• List, view, and run integrations
• And more

The server is designed to be deployed on Smithery for easy access via WebSocket connections.

Tools

Workspace Tools

• list_workspaces: Retrieve all available workspaces
• get_workspace: Get details about a specific workspace

Connection Tools

• list_connections: List connections in a workspace
• get_connection_details: Get details about a specific connection
• test_connection: Test a connection

Integration Tools

• list_integrations: List integrations in a workspace
• get_integration: Get details about a specific integration
• run_integration: Execute an integration

Deployment

Prerequisites

• Docker installed on your local machine
• Access to Smithery deployment
• Skyvia API token

Local Testing

To test the MCP server locally:

1. Install dependencies:

   pip install -r requirements.txt
   

2. Set environment variables:

   export SKYVIA_API_TOKEN="your-api-token-here"
   

3. Run the server:

   python main.py
   

4. Use an MCP client like MCP Inspector to test:

   npx @modelcontextprotocol/inspector python main.py
   

Smithery Deployment

1. Add the server to Smithery
2. Configure the server with your Skyvia API token
3. Deploy and test the server through Smithery’s interface

Configuration

The server requires a Skyvia API token to authenticate with the Skyvia API. This token should be provided via the SKYVIA_API_TOKEN environment variable.

Usage Examples

Here’s how to use some of the tools:

List Workspaces

// Example MCP client code
const result = await client.useToolWithMCP("list_workspaces", {});
console.log(result);

List Integrations

// Example MCP client code
const result = await client.useToolWithMCP("list_integrations", {
  workspace_id: 123 // Replace with your workspace ID
});
console.log(result);

Run an Integration

// Example MCP client code
const result = await client.useToolWithMCP("run_integration", {
  workspace_id: 123, // Replace with your workspace ID
  integration_id: 456 // Replace with your integration ID
});
console.log(result);

Troubleshooting

“Server not initialized” Error

If you receive a “Server not initialized” error when trying to use the server tools, this typically indicates that the server did not properly initialize due to a missing or invalid API token. To resolve this:

1. Verify your Skyvia API token is valid and has the necessary permissions.
2. Make sure the SKYVIA_API_TOKEN environment variable is properly set:
   • For local testing, use export SKYVIA_API_TOKEN="your-token-here"
   • For Smithery deployment, ensure the token is correctly set in your Smithery profile

Diagnosing Connection Issues

The repository includes a dedicated diagnostic script for testing connectivity:

# Set your API token
export SKYVIA_API_TOKEN="your-token-here"

# Optionally set your Smithery API key (if using a different one)
export SMITHERY_API_KEY="your-smithery-api-key"

# Run the diagnostic script
python test_smithery_connection.py

This script will:

• Test direct connection to a locally running server
• Test connection to the Smithery-deployed server
• Provide detailed logs and troubleshooting guidance
• Compare results to help isolate if the issue is with Smithery configuration or the server itself

Common Issues

1. Missing Dependencies: Ensure you’ve installed all required packages with pip install -r requirements.txt
2. Invalid API Token: The token may be expired or have insufficient permissions
3. Smithery Profile Issue: If using Smithery, check that your profile has the correct configuration with the API token
4. Networking Issues: If behind a corporate firewall, check if websocket connections are allowed

Contributing

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

License

MIT