Look 4 Fundings - EU Funding Crawler MCP Server

A Model Context Protocol (MCP) server that provides tools for searching EU funding opportunities through the official EU Funding & Tenders Portal API. This server is designed to work with Le Chat and other MCP-compatible clients.

Features

• EU Funding Search: Search for active EU funding opportunities by keyword
• Comprehensive Data: Get detailed information including title, summary, deadline, budget, and status
• Official API Integration: Uses the official EU Funding & Tenders Portal REST API
• Le Chat Integration: Compatible with Le Chat's MCP connector system

Prerequisites

• Python 3.12 or higher
• uv package manager

Installation

1. Clone or download this repository
2. Install dependencies using uv:

uv sync

Running the Server

Method 1: Direct Python Execution

Run the server directly with Python:

python main.py

The server will start on http://localhost:3000 and use the streamable-http transport for Le Chat compatibility.

Method 2: Using LocalTunnel for External Access

To expose your local server to the internet (required for Le Chat integration), use lt (LocalTunnel):

1. Install LocalTunnel globally:

npm install -g localtunnel

2. In one terminal, start the MCP server:

python main.py

3. In another terminal, expose the server using LocalTunnel:

lt --port 3000

This will provide you with a public URL like https://legal-bugs-chew.loca.lt that you can use to connect to your MCP server.

Le Chat Integration

To connect this MCP server to Le Chat:

1. Start your MCP server using one of the methods above
2. If using LocalTunnel, note the provided URL (e.g., https://legal-bugs-chew.loca.lt)
3. In Le Chat, go to the connectors section
4. Add a new MCP connector with the URL: https://your-tunnel-url.loca.lt/mcp
5. The server will be available for use in Le Chat

Available Tools

search_eu_fundings

Searches for EU funding opportunities by keyword and returns detailed funding information.

Parameters:

• keyword (str): The search keyword (e.g., "AI", "machine learning", "renewable energy") - defaults to "AI"
• page_size (int): Number of results to return per page - defaults to 20

Returns:

• List[PublicFunding]: A list of PublicFunding objects containing:
  • title (str): Title of the funding opportunity
  • url (str): Direct link to the funding page
  • summary (str): Summary/objective of the funding
  • deadline (str): Application deadline
  • status (str): Current status of the funding
  • budget (str): Budget information with currency

Example:

Input: keyword="artificial intelligence", page_size=10
Output: List of 10 EU funding opportunities related to AI

Note: Returns an empty list if no results are found or if an error occurs during the API request.

Development

This project uses:

• FastMCP: For building the MCP server
• Streamable HTTP: For Le Chat compatibility
• Python 3.12+: For modern Python features
• Pydantic: For data validation and serialization
• Requests: For HTTP API calls to the EU Funding Portal
• EU Funding & Tenders Portal API: Official REST API for funding data

Project Structure

• main.py: MCP server setup and tool definitions
• crawler.py: EUFundingCrawler class for API interactions
• type.py: Pydantic models for data structures
• pyproject.toml: Project dependencies and configuration

Troubleshooting

• Port already in use: If port 3000 is occupied, you can modify the port in main.py and update the LocalTunnel command accordingly
• LocalTunnel issues: Make sure LocalTunnel is installed globally and the tunnel URL is accessible
• Le Chat connection: Ensure the MCP endpoint URL includes /mcp at the end
• API errors: The crawler uses the official EU API which may have rate limits or temporary unavailability
• No results found: Try different keywords or check if the API is responding correctly

License

This project is open source and available under the MIT License.