home / mcp / cdata sync mcp server
Provides MCP tools to manage CData Sync data connections, jobs, and ETL through dual transport (stdio and HTTP).
Configuration
View docs{
"mcpServers": {
"cdatasoftware-cdata-sync-mcp-server": {
"url": "http://your-server:3000/mcp/v1",
"headers": {
"MCP_HTTP_PATH": "/mcp/v1",
"MCP_HTTP_PORT": "3000",
"CDATA_BASE_URL": "http://localhost:8181/api.rsc",
"CDATA_PASSWORD": "pass",
"CDATA_USERNAME": "user",
"CDATA_WORKSPACE": "workspace-uuid",
"CDATA_AUTH_TOKEN": "YOUR_TOKEN",
"MCP_TRANSPORT_MODE": "http"
}
}
}
}You can run the CData Sync MCP Server to expose REST API controls as MCP tools, enabling AI assistants to manage data synchronization jobs, connections, and ETL tasks. It supports both local (stdio) usage and remote (HTTP) deployments, giving you flexible options for development and production workloads.
Use the MCP Server to manage CData Sync entities through either a local stdio setup for desktop use or a remote HTTP setup for web-based integrations. You can list, read, create, update, and delete connections, jobs, tasks, transformations, users, and more. Real-time updates via streaming let you monitor job progress and tool activity as it happens.
For development and testing, you can run both transports simultaneously to compare behaviors in desktop and remote environments. In production, prefer a dedicated HTTP deployment or a combined approach based on your infrastructure needs.
Prerequisites: Node.js 18+ and a running CData Sync instance. Decide whether you will run the MCP Server locally with stdio or deploy it remotely with HTTP. Ensure you have a workspace UUID if you want to scope operations.
Step 1: Install dependencies after obtaining the MCP server package.
Step 2: Build the project to produce the runtime artifacts.
Step 3: Configure environment variables to connect to your CData Sync instance and set the transport mode. You will typically provide the base URL, an auth token, and an optional workspace.
Step 4: Start the desired transport mode: stdio for local desktop usage, HTTP for remote deployments, or both for development testing.
The server supports these environment variables to configure connectivity and behavior. Typical values are provided as examples and should be replaced with your actual credentials and endpoints.
When using stdio, you define a command that runs the runtime index and pass necessary environment variables to shape the runtime behavior.
When using HTTP, you run a server that exposes the MCP endpoints and handles streaming updates through Server-Sent Events.
{
"mcpServers": {
"cdata-sync-server": {
"command": "node",
"args": ["/absolute/path/to/cdata-sync-mcp-server/dist/index.js"],
"env": {
"MCP_TRANSPORT_MODE": "stdio",
"CDATA_AUTH_TOKEN": "your-token-here",
"CDATA_BASE_URL": "http://localhost:8181/api.rsc",
"CDATA_WORKSPACE": "your-workspace-uuid-here",
"DISABLE_SSE": "true"
}
}
}
}List, count, get details, or test connections
Create, update, or delete connections
List tables in a connection
Get table schema information
List, count, get details, status, history, or logs
Create, update, or delete jobs
Run a sync job immediately
Stop a running job
Run custom SQL queries
List, count, or get task details
Create, update, or delete tasks
List, count, or get transformation details
Create, update, or delete transformations
List, count, or get user details
Create or update users
List, count, or get request log details
Delete request logs
List or count execution history records
List certificates
Create certificates
Get or update server configuration