gradio -hack

This MCP server delivers a Gradio-based interface for two separate tracks focused on France unemployment (chômage) data. Track 1 provides AI agents with tools to search for and analyze job offers, enabling users to query listings, compare requirements, and extract insights from listings through the MCP-enabled Gradio UI. Track 2 demonstrates a multi-agent setup where agents collaborate to match a user profile and search criteria with relevant job offers, leveraging the MCP tooling layer to coordinate tool usage, summarize results, and refine searches. To use it, run the server (see installation steps) and open the Gradio interface that starts up as part of the python module execution. Interact with the agents through natural language prompts, and rely on the integrated tools (search, filter, summarize) to gather and reason about job offers.

Within the tools, you can perform actions such as:

• Searching job boards or sources for offers matching criteria.
• Filtering results by location, salary, contract type, and experience level.
• Analyzing offer texts to extract key requirements and responsibilities.
• Coordinating multiple agents in Track 2 to align on profile suitability and preferred searches.

The MCP layer coordinates tool usage across agents, returns structured results, and allows you to iteratively refine queries based on previous responses. This makes it possible to build complex, goal-driven workflows for job-market analysis using the Gradio UI provided by the project.

Prerequisites:

• Python 3.10–3.13
• Git
• (Optional) a virtual environment tool (venv, conda, etc.)

1. Clone the repository

git clone https://github.com/AubinSeptier/gradio-mcp-hack.git
cd gradio-mcp-hack

2. Create and activate a virtual environment

On Linux/macOS:

python -m venv .venv
source .venv/bin/activate

On Windows:

python -m venv .venv
.\.venv\Scripts\activate

3. Install dependencies

pip install -e .

4. Run the MCP server

python -m france_chomage_mcp_server.app

If you prefer development dependencies, install with extras:

pip install -e ".[dev]"

Notes:

• Ensure you are using Python 3.10–3.13 as required.
• The project includes pre-commit hooks; you can install and run them as described in the repository.

Tips and notes:

• The MCP server is organized into two Gradio tracks; Track 1 focuses on single-agent tool usage for job offer search/analysis, while Track 2 demonstrates multi-agent coordination to match profiles with search criteria.
• If you encounter environment-related errors, verify that the virtual environment is activated and dependencies are installed via pip install -e .
• If the Gradio UI does not load, check that Python is launching the module correctly and that any required data sources for job offers are accessible.
• You can customize environment variables as needed; for example, set GRADIO_API_SERVER for remote hosting or adjust logging with standard Python logging configurations.