DeBank

You deploy the DeBank MCP Server to enable natural-language queries against the DeFi data API through an MCP client. This server bridges DeBank data with AI assistants, letting you check balances, tokens, NFTs, transaction history, and more through conversational queries while keeping control over keys and configuration.

How to use

Use an MCP client to connect to the DeBank MCP Server via a local stdio integration. Start the server from your development environment and point the client at the local Python module that runs the MCP server. You can then ask questions about wallets, tokens, protocols, and DeFi activity in natural language, and the server will fetch data from DeBank using your API key.

How to install

Prerequisites ensure you have Python 3.10 or newer and an active DeBank Cloud API key. You also need a client that supports MCP, such as Claude Desktop or another MCP-compatible interface.

# Clone the repository
git clone https://github.com/caiovicentino/debank-mcp-server.git
cd debank-mcp-server

# Install the dependencies
pip install -e .

Additional setup and usage notes

Configure your DeBank API key in an environment file and ensure the client can access the module paths for the server. The recommended setup uses a local stdio integration that runs the MCP server as a Python module.

To test that the server is ready, run the Python snippet to verify the MCP server module can be imported and executed.

Configuration and security

Create a .env file at the project root and store your API key securely. Do not share this key, and add the file to your .gitignore to prevent accidental commits.

Example of a minimal environment configuration to run the MCP server locally via Claude Desktop:

{
  "mcpServers": {
    "debank": {
      "command": "python",
      "args": [
        "-m",
        "mcp_server_debank.server"
      ],
      "cwd": "/path/to/debank-mcp-server",
      "env": {
        "DEBANK_ACCESS_KEY": "YOUR_API_KEY_HERE",
        "PYTHONPATH": "/path/to/debank-mcp-server/src"
      }
    }
  }
}

Troubleshooting common issues

If the server fails to start, verify that your API key is set in the environment and that the working directory (cwd) points to the correct project location. Ensure the PYTHONPATH includes the src directory so Python can locate the MCP modules.

Project structure and runtime details

The MCP server exposes tools for core data sources, portfolio analytics, and advanced features, all designed to work with a variety of DeBank endpoints. Start the server as a Python module from your configured environment and connect your MCP client to the provided stdio interface.

Examples of usage

After configuration, you can pose natural-language questions to your MCP client, such as asking for wallet balances, token information, NFT counts, DeFi protocol positions, or transaction history, and receive structured responses backed by DeBank data.

Available tools

Chains

List all supported blockchains (93+ chains) to query data across networks.

Protocols

Provide information about DeFi protocols including Total Value Locked (TVL).

Tokens

Fetch token prices, metadata, and holders.

Balance

Compute the total balance of wallets across chains.

User Tokens

Query token holdings for a user with pagination support.

User NFTs

Retrieve NFT collections with metadata for a user.

User Protocols

Show user DeFi positions with simple or complex views.

User History

Access transaction history for a wallet.

User Approvals

Analyze security of token approvals.

Net Curve

Track 24h value trends.

Pool Info

Analyze liquidity pool analytics.

Transaction Simulation

Simulate transactions before sending.

Gas Prices

Get real-time gas prices by tier.

Account Units

Monitor API usage units.

User Social

Placeholder for future OAuth-related features.