home / mcp / yingdao mcp server
影刀RPA MCP Server
Configuration
View docs{
"mcpServers": {
"ying-dao-yingdao_mcp_server": {
"url": "http://localhost:3000/sse",
"headers": {
"RPA_MODEL": "openApi",
"ACCESS_KEY_ID": "YOUR_ACCESS_KEY_ID",
"ACCESS_KEY_SECRET": "YOUR_ACCESS_KEY_SECRET"
}
}
}
}影刀RPA MCP Server enables RPA automation capabilities to be invoked by MCP-compatible clients, providing both local/embedded and API-access modes. It helps you connect RPA workflows to AI agents and other MCP hosts, with options for running locally or via open API authentication, and supports both SSE and Stdio server communication patterns.
You will connect an MCP client to YingDao RPA MCP Server to run and manage RPA applications. Choose between two modes: local execution within your environment or an open API integration for enterprise use. You can start an SSE-based server endpoint for real-time streaming or a Stdio-based local process for direct host integration. In your MCP client, specify the server with its URL for SSE or provide the local stdio command to spawn the server process.
Prerequisites: you need Node.js and npm installed on your machine.
1) Clone the MCP server repository.
2) Install dependencies.
3) Build the SSE server.
4) Start the SSE server and configure your MCP client to point to the server URL.
Exact steps shown in the configuration sections below describe how to run the server in both Stdio and SSE modes.
Stdio server configuration enables a local, tightly coupled MCP integration. Use the following command and arguments to run the server via a package runner. Ensure your environment variables match your chosen mode.
{
"mcpServers": {
"YingDao RPA MCP Server": {
"command": "npx",
"args": ["-y", "yingdao-mcp-server"],
"env":{
"RPA_MODEL":"openApi",
"ACCESS_KEY_ID":"{your_access_key_id}",
"ACCESS_KEY_SECRET":"{your_access_key_secret}"
}
}
}
}Protect your API keys and credentials. Use the openApi mode only in trusted enterprise environments and rotate keys regularly.
Query the parameters of a given RPA application to understand its configuration and requirements.
Retrieve the list of available RPA applications, with support for pagination in the open API mode.
Execute a selected RPA application within the MCP environment.
Upload files to the RPA platform when using the open API mode.
Start an RPA job to run automation tasks asynchronously.
Check the status and progress of a running or completed RPA job.
Query the list of scheduled RPA robots or client instances.