🧠 FastMCP SSE Server – Research Paper Agent
LocalAbout 🧠 FastMCP SSE Server – Research Paper Agent
This FastMCP remote server helps researchers search arXiv papers, retrieve saved paper information, and prepare topic reports. It is intended for Claude, GPT, and other MCP clients that support SSE.
Key features
- 01Search and save arXiv papers by topic
- 02Retrieve details for a saved paper
- 03Read summaries of papers by topic
- 04List saved topic folders
- 05Generate topic reports from a prompt template
Use cases
- 01Discover papers on a research topic
- 02Retrieve information about a saved paper
- 03Review summaries grouped by topic
- 04Draft a structured topic report
MCP support
Exposes paper search and retrieval tools, saved-paper resources, and a report-generation prompt to MCP clients over SSE.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEtitle: MCP_Research_Server app_file: main.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
To test if it’s working, simply visit the link above — you’ll see a plain text confirmation.
🚀 Features
search_papers(topic): Search and save top arXiv papers by topicextract_info(paper_id): Retrieve paper details from stored JSONget_topic_papers(topic): Read summaries for all papers in a topicget_available_folders(): List all saved topic folders- Prompt template for Claude to generate full topic reports
🧑💻 Project Structure
.
├── main.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: A fast Python package manager
- Render.com (for deployment)
🛠️ Local Setup (Optional)
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 main.py
The server will run on localhost:8001/sse.
☁️ Deploy on Render.com (Docker)
- Push this project to your GitHub
- Create a new web service on Render
- Use the following settings:
- Environment: Docker
- Port: 8001
- Start command: (leave blank – handled in Dockerfile)
- Deploy 🚀
Render will give you a URL like:
https://your-app-name.onrender.com/sse
To run locally in Docker:
docker run -p 8001:8001 <your-image-name> python main.py
🧪 Test with MCP Inspector
Install and run:
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.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.