automatisch
MCP Server for Automatisch workflow automation platform
claude mcp add --transport stdio milisp-automatisch-mcp-server npx -y automatisch-mcp-server \ --env AUTOMATISCH_API_KEY="your-api-key" \ --env AUTOMATISCH_BASE_URL="http://localhost:3001"
How to use
Automatisch MCP Server exposes an MCP interface to control and observe Automatisch workflow automation. It provides tools for Workflow Management (list, get, create, update, delete, test), Connection Management (list and create app connections), Execution Monitoring (list executions), and App Discovery (list available apps). You can interact with these capabilities via the MCP endpoints exposed by the server or through Claude/Desktop integrations that load the server as a tool. To use it, register the MCP server under your preferred client (for example, Claude Desktop) and supply Automatisch base URL and API key if you have one. In standalone mode, you can start the server and issue commands through the MCP-compatible interface to manage workflows, connections, and view execution history. The server also includes example commands and prompts to help you query available apps, create workflows, or inspect execution status.
How to install
Prerequisites:
- Node.js 18+ installed on your machine
- Access to an Automatisch instance (self-hosted or hosted) with API access
Installation steps:
- Clone or download the MCP server repository:
git clone https://github.com/milisp/automatisch-mcp-server.git
cd automatisch-mcp-server
- Install dependencies:
npm install
- Build the project (if applicable):
npm run build
- Configure environment variables (example):
# Automatisch instance URL (default: http://localhost:3001)
export AUTOMATISCH_BASE_URL="http://your-automatisch-instance:3001"
# API key for authentication (optional but recommended)
export AUTOMATISCH_API_KEY="your-api-key"
- Run the server in development or production mode:
# Development mode
npm run dev
# Production / standard start
npm start
- (Optional) If you use Claude Desktop or another MCP client, register the MCP server with the client's configuration using the provided command and environment variables.
Additional notes
Notes:
- The server communicates with Automatisch via base URL and optional API key. Keep the API key secure and restrict access to trusted networks.
- If you change AUTOMATISCH_BASE_URL, ensure the URL is reachable from the host running the MCP server.
- For debugging, you can enable NODE_ENV=development to get verbose logs.
- When integrating with Claude Desktop, ensure the MCP server is registered under the expected name (e.g., automatisch) to use the provided example configuration.
- Common issues include connectivity to Automatisch, invalid API keys, and misconfigured environment variables. Check Automatisch logs and network access if problems arise.
Related MCP Servers
sudocode
Lightweight agent orchestration dev tool that lives in your repo
mcp-atlassian
MCP server connecting AI assistants with Jira & Confluence for smart project management.
browser-use -client
A MCP client for browser-use
notebooklm
MCP server for NotebookLM - Let your AI agents (Claude Code, Codex) research documentation directly with grounded, citation-backed answers from Gemini. Persistent auth, library management, cross-client sharing. Zero hallucinations, just your knowledge base.
autonomo
Tired of 'it works' lies? Autonomo MCP makes your AI prove it—on real hardware, right in your editor.
prometheus
A Model Context Protocol (MCP) server implementation that provides AI agents with programmatic access to Prometheus metrics via a unified interface.