Valyu

The Valyu MCP Server lets AI models fetch rich context from Valyu’s API, including full search coverage over Wikipedia, arXiv papers, and the web. It’s designed to power you with high-quality, AI-friendly context in your MCP-enabled workflows.

How to use

You use this server by configuring an MCP client to load the Valyu MCP tool. Once the server is running, open your MCP client (for example, Claude Desktop) and verify the valyu_mcp tool is available in the MCP tool list. Run a query in your client to fetch current context from Valyu, such as a search for recent reinforcement learning papers. The MCP server handles the API integration and returns results that your AI model can natively consume.

How to install

Prerequisites you need before you begin are Python 3.10 or newer, Claude Desktop (latest version), and a Valyu API Key. Check your Python version with python --version.

python --version

Option 1: Automated Setup (Recommended)

Clone the repository and run the setup script. The setup creates and activates a virtual environment, installs dependencies, prompts for your Valyu API key, creates the .env file, and provides Claude Desktop configuration steps.

git clone https://github.com/ValyuNetwork/valyu-mcp.git
cd valyu-mcp

chmod +x setup.sh
./setup.sh

Option 2: Manual Setup

Clone the repository, create and activate a virtual environment, install dependencies, and create the environment file.

git clone https://github.com/ValyuNetwork/valyu-mcp.git
cd valyu-mcp

python -m venv .venv
source .venv/bin/activate  # macOS/Linux
.venv\Scripts\activate    # Windows

pip install -r requirements.txt

Create the environment file with your API key.

echo "VALYU_API_KEY=your-api-key-here" > .env

Open Claude Desktop configuration to add the MCP server entry. You will place a server entry under the mcpServers section that points to the Python interpreter in your virtual environment and the Valyu MCP script.

{
  "mcpServers": {
    "valyu_mcp": {
      "command": "/ABSOLUTE/PATH/TO/.venv/bin/python",
      "args": ["-u", "/ABSOLUTE/PATH/TO/valyu-mcp.py"],
      "env": {
        "VALYU_API_KEY": "your-api-key-here"
      }
    }
  }
}

Start the MCP server to verify it runs. You can test it interactively by running the script directly.

python valyu-mcp.py

Configure Claude Desktop to recognize the new tool and perform a test query, such as asking for latest papers on reinforcement learning.

Additional setup notes

If you encounter issues, check logs for clues. On macOS you can tail the MCP logs with tail -n 20 -F ~/Library/Logs/Claude/mcp*.log. On Windows, inspect the log file at %APPDATA%\Claude\Logs\mcp_valyu-mcp.log.

Common problems include ensuring the virtual environment is active when running the script and verifying the paths in Claude Desktop’s config file. If no results are returned from the Valyu API, confirm that your API key is valid and has credits.

Security and access

Keep your Valyu API key secure. Do not share the key or commit it to version control. The key is provided by Valyu Exchange and is used to access the search and context capabilities through the MCP server.

Notes and credits

You can sign up at Valyu Exchange to obtain an API key and, if available, claim introductory credits to test the MCP server features.

Troubleshooting

Module or runtime errors typically indicate issues with the virtual environment activation or missing dependencies. Ensure you are running commands in the correct environment and with the proper path references. If the server fails to start, re-check the command and script paths configured in Claude Desktop.