OpenProject Integration Server connects AI agents to self-hosted OpenProject instances, enabling comprehensive project and task management automation through the Model Context Protocol. Key capabilities: - Full project lifecycle management including creation, retrieval, updates, deletion, and paginated listing of OpenProject projects - Complete work package (task) management with support for creating, retrieving, updating, and deleting tasks, plus filtered task lists by project - HTTP-based MCP transport built with Express for accessible remote client connections via the `/mcp` endpoint - Cloud deployment support for platforms like Smithery and Netlify with Docker containerization and production-ready configuration - Real-time operation support through SSE bridging capabilities for live connectivity between AI agents and OpenProject instances

[](https://smithery.ai/server/@jessebautista/mcp-openproject-smithery)

MCP Server for OpenProject (Smithery Edition)

This project provides a Model Context Protocol (MCP) server for OpenProject, designed for deployment and use with Smithery. It exposes a set of tools for interacting with a self-hosted OpenProject instance, and is compatible with Smithery's HTTP transport and tool listing requirements.

About this MCP Server

Implemented OpenProject Tools

Prerequisites

HTTP Entrypoint and Local Development

This MCP server now uses Express and the official MCP Streamable HTTP transport. The server is accessible at http://localhost:8000/mcp after running npm run dev or npm start.

Entrypoint code (see src/index.ts):

import express from "express";
import { StreamableHTTPServerTransport } from "@modelcontextprotocol/sdk/server/streamableHttp.js";
const app = express();
app.use(express.json());
const mcpServer = setupMCPServer();
const transport = new StreamableHTTPServerTransport({ sessionIdGenerator: undefined });
mcpServer.connect(transport);
app.all("/mcp", (req, res) => {
  transport.handleRequest(req, res, req.body);
});
app.get("/", (_req, res) => res.send("MCP OpenProject server is running!"));
const PORT = process.env.PORT || 8000;
app.listen(PORT, () => {
  console.log(MCP server running on port ${PORT});
});

Running Locally

1. Install dependencies:

   npm install
   

   npm run build
   

   npm run dev
   # or for production
   npm start
   

Docker Usage

Build and run as before:

docker build -t mcp-openproject-smithery .
docker run --rm -p 8000:8000 --env-file .env mcp-openproject-smithery

Smithery Deployment

This project is ready for Smithery. The /mcp endpoint is compatible with Smithery's HTTP transport requirements.

See your live Smithery server here: [](https://smithery.ai/server/@jessebautista/mcp-openproject-smithery)

Local Testing with MCP Inspector

You can test your MCP server locally using the MCP Inspector:

npx @modelcontextprotocol/inspector npx mcp-remote@next http://localhost:8000/mcp

Open the Inspector URL (usually http://localhost:6274) in your browser to interact with your tools.

Project Structure

License

MIT

---

This project is maintained for use with Smithery and the Model Context Protocol. For more information, see: