| | --- |
| | title: MCP_Research_Server |
| | app_file: research_server.py |
| | sdk: gradio |
| | sdk_version: 5.31.0 |
| | --- |
| | # 🧠 FastMCP SSE Server – Research Paper Agent |
| |
|
| | This project is a deployable **MCP-compatible remote server** built using the `FastMCP` framework. It exposes tools and resources for: |
| |
|
| | - Searching academic papers on **arXiv** |
| | - Extracting information about saved papers |
| | - Generating structured prompts for Claude or other LLM agents |
| |
|
| | It is designed to work with **Claude, GPT, or any MCP client** that supports `SSE` transport. |
| |
|
| | --- |
| |
|
| | ## 🌐 Live Server |
| |
|
| | ✅ **MCP server is running here:** |
| | **Tool URL (SSE):** [`https://mcp-server-vs1x.onrender.com/sse`](https://mcp-server-vs1x.onrender.com/sse) |
| |
|
| | To test if it’s working, simply visit the link above — you’ll see a plain text confirmation. |
| |
|
| | <img width="496" alt="image" src="https://github.com/user-attachments/assets/90fc6c84-a7af-4f73-9e7e-fce36f7234e5" /> |
| |
|
| |
|
| | --- |
| |
|
| | ## 🚀 Features |
| |
|
| | - `search_papers(topic)`: Search and save top arXiv papers by topic |
| | - `extract_info(paper_id)`: Retrieve paper details from stored JSON |
| | - `get_topic_papers(topic)`: Read summaries for all papers in a topic |
| | - `get_available_folders()`: List all saved topic folders |
| | - **Prompt template** for Claude to generate full topic reports |
| |
|
| | --- |
| |
|
| | ## 🧑💻 Project Structure |
| |
|
| | ```bash |
| | . |
| | ├── research_server.py # Main FastMCP server |
| | ├── Dockerfile # For deployment on Render |
| | ├── pyproject.toml # Python project setup (required by uv) |
| | ├── uv.lock # Dependency lock file (required by uv) |
| | ├── papers/ # Local storage for downloaded paper info |
| | ``` |
| |
|
| | --- |
| |
|
| | ## 📦 Requirements |
| |
|
| | - Python 3.11+ |
| | - [uv](https://github.com/astral-sh/uv): A fast Python package manager |
| | - [Render.com](https://render.com) (for deployment) |
| |
|
| | --- |
| |
|
| | ## 🛠️ Local Setup (Optional) |
| |
|
| | ```bash |
| | git clone https://github.com/YOUR_USERNAME/mcp-sse-server.git |
| | cd mcp-sse-server |
| | |
| | # Run with uv (you must have uv installed) |
| | uv pip install --system . |
| | uv run research_server.py |
| | ``` |
| |
|
| | The server will run on `localhost:8001/sse`. |
| |
|
| | --- |
| |
|
| | ## ☁️ Deploy on Render.com (Docker) |
| |
|
| | 1. Push this project to your GitHub |
| | 2. Create a new web service on Render |
| | 3. Use the following settings: |
| | - **Environment:** Docker |
| | - **Port:** 8001 |
| | - **Start command:** (leave blank – handled in Dockerfile) |
| | 4. Deploy 🚀 |
| |
|
| | Render will give you a URL like: |
| | ``` |
| | https://your-app-name.onrender.com/sse |
| | ``` |
| |
|
| | **To run locally in Docker:** |
| |
|
| | ```bash |
| | docker run -p 8001:8001 <your-image-name> python research_server.py |
| | ``` |
| |
|
| | --- |
| |
|
| | ## 🧪 Test with MCP Inspector |
| |
|
| | Install and run: |
| |
|
| | ```bash |
| | npx @modelcontextprotocol/inspector |
| | ``` |
| |
|
| | In the web UI: |
| | - **Transport:** SSE |
| | - **URL:** `https://mcp-server-vs1x.onrender.com/sse` |
| |
|
| | You’ll now be able to call the tools and test them live using Claude or your own chatbot. |
| |
|
| | --- |
| |
|
| | ## 📚 Credits |
| |
|
| | Built as part of the DeepLearning.AI Claude Agent Systems course. |
| |
|
