astron-rpa
Agent-ready RPA suite with out-of-the-box automation tools. Built for individuals and enterprises.
claude mcp add --transport stdio iflytek-astron-rpa docker run -i iflytek/astron-rpa:latest
How to use
AstronRPA is an enterprise-grade RPA desktop automation stack that combines a frontend Electron client, a Python/Java backend, and a suite of automation components. The MCP deployment described here uses Docker to run the Astron-RPA server stack with Casdoor for authentication and the RPA engine for workflow execution. Once started, you can access the system's authentication endpoint and the admin UI, and you can trigger automation workflows, API calls, and scheduling through the integrated MCP services. The server supports multi-channel triggers, including direct execution, scheduling, and API-based activations, enabling seamless integration with other systems in your enterprise.
How to install
Prerequisites:
- Docker and Docker Compose installed on your host
- Basic familiarity with CLI operations
- Network access to required services (e.g., Casdoor, database if used by the deployment)
Installation steps:
- Clone the repository and navigate to the deployment directory:
git clone https://github.com/iflytek/astron-rpa.git
cd astron-rpa
- Open the docker deployment folder and prepare environment:
cd docker
# Copy the example environment file to actual env file
cp .env.example .env
- Configure environment variables in .env, notably the Casdoor endpoint and service addresses:
# Example configuration (adjust to your environment)
CASDOOR_EXTERNAL_ENDPOINT="http://YOUR_SERVER_IP:8000"
# Add or modify other necessary environment variables as documented in QUICK_START.md
- Start the services using Docker Compose (recommended quick-start path):
docker compose up -d
- Verify services are running and accessible:
docker compose ps
- Access the authentication/login flow and admin endpoints as described in the project’s Deployment Guide. For production hardening, refer to the Deployment Guide in ./docker/QUICK_START.md.
Note: The project also provides a direct source/build workflow and client deployments, but the Docker path described above is the quickest route to run the MCP server stack.
Additional notes
Tips and notes:
- The default ports referenced are 32742 for the RPA auth endpoint and 8000 for Casdoor. If you modify configuration, adjust these values accordingly.
- Ensure Casdoor and any required databases/services are reachable from the host running Docker.
- When using the docker deployment, environment variables in .env should reflect your infrastructure (endpoints, IPs, and security settings).
- This server stack expects a Docker-based deployment; if you prefer a direct Python/Java/Node deployment, consult the repository’s BUILD_GUIDE.md for alternate install paths.
- If you encounter authorization errors, verify that CASDOOR_EXTERNAL_ENDPOINT is reachable and that Casdoor is deployed correctly in the stack.
Related MCP Servers
gpt-researcher
An autonomous agent that conducts deep research on any data using any LLM providers.
astron-agent
Enterprise-grade, commercial-friendly agentic workflow platform for building next-generation SuperAgents.
magic
Super Magic. The first open-source all-in-one AI productivity platform (Generalist AI Agent + Workflow Engine + IM + Online collaborative office system)
better-chatbot
Just a Better Chatbot. Powered by Agent & MCP & Workflows.
flow-like
Flow-Like: Strongly Typed Enterprise Scale Workflows. Built for scalability, speed, seamless AI integration and rich customization.
lihil
2X faster ASGI web framework for python, offering high-level development, low-level performance.