uipath
UiPath MCP Server implementation for on-premise Automation Suite & Orchestrator
claude mcp add --transport stdio javaos74-uipath-mcp-server uvx uipath-mcp-server \ --env DB_PATH="Database file path (default: database/mcp_servers.db)" \ --env API_HOST="Server host (default: 0.0.0.0)" \ --env API_PORT="Server port (default: 8000)" \ --env LOG_LEVEL="Logging level (default: INFO)" \ --env SECRET_KEY="JWT secret key (required for production)" \ --env TOOL_CALL_TIMEOUT="UiPath tool execution timeout in seconds (default: 600)"
How to use
This UiPath MCP Server provides a multi-tenant API plus a web UI to manage multiple UiPath endpoints. It runs a Python backend (FastAPI/Starlette) with authentication and a React-based frontend for configuring UiPath connections, creating MCP servers, and defining tools that map to UiPath processes. The server exposes MCP protocol endpoints for real-time communication via SSE and HTTP streaming, so MCP clients can subscribe to updates and post commands. To get started, install the backend and frontend, configure UiPath authentication (PAT or OAuth), create an MCP server, and then add tools that point to UiPath processes. The UI also exposes endpoints to list folders and processes in UiPath, and supports token-based authentication for server access.
How to install
Prerequisites
- Python 3.11+ (backend)
- Node.js 18+ (frontend)
- uv (Python package manager)
Installation steps
-
Clone the repository git clone https://github.com/your-org/uipath-mcp-server.git cd uipath-mcp-server
-
Setup backend dependencies cd backend python -m venv venv source venv/bin/activate # On Windows use: venv\Scripts\activate pip install -r requirements.txt
-
Setup frontend dependencies cd frontend npm install
-
Environment configuration Create a .env file in backend with required variables:
- API_HOST=0.0.0.0
- API_PORT=8000
- DB_PATH=database/mcp_servers.db
- SECRET_KEY=your-secret-key
- TOOL_CALL_TIMEOUT=600
- LOG_LEVEL=INFO
-
Run in development
- Backend: cd backend && uv run python -m src.main
- Frontend: cd frontend && npm run dev
-
Production build (optional)
- Build frontend and run backend together as described in the docs: ./build.sh # Build frontend (outputs to backend/static/) cd backend && uv run python -m src.main
-
Docker (optional)
- Build frontend first: ./build.sh
- Build Docker image: ./docker-build-simple.sh
- Run: docker-compose up -d
Note: Replace placeholders with your actual UiPath URLs, client credentials, and tokens as needed.
Additional notes
Tips and notes:
- The default admin account is created when the database is first initialized: username admin, password admin. Change this immediately after first login.
- UiPath integration supports PAT (for UiPath Cloud) and OAuth 2.0 (for On-Premise). Configure the chosen method in the Settings page.
- SSL handling: self-signed certificates are supported for On-Prem installations; SSL verification is disabled for non-cloud URLs.
- The MCP protocol endpoints require an Authorization header with a server API token: Authorization: Bearer <server-api-token>.
- The system supports multiple MCP servers, each with its own set of tools and UiPath configurations.
- If you deploy via Docker, the image tag reflects the version in backend/pyproject.toml (currently 0.1.0). See DOCKER.md for detailed docker deployment guidance.
Related MCP Servers
mcp-vegalite
MCP server from isaacwasserman/mcp-vegalite-server
github-chat
A Model Context Protocol (MCP) for analyzing and querying GitHub repositories using the GitHub Chat API.
nautex
MCP server for guiding Coding Agents via end-to-end requirements to implementation plan pipeline
pagerduty
PagerDuty's official local MCP (Model Context Protocol) server which provides tools to interact with your PagerDuty account directly from your MCP-enabled client.
futu-stock
mcp server for futuniuniu stock
mcp -boilerplate
Boilerplate using one of the 'better' ways to build MCP Servers. Written using FastMCP