Crypto Trending

You deploy this MCP server to track the latest trending cryptocurrencies from CoinGecko and enable real-time analysis through an MCP client. It exposes a simple local workflow that integrates with a client like Claude Desktop, letting you ask natural language questions about trending coins and have the data parsed and analyzed automatically.

How to use

To use this server, you run it locally and connect it to an MCP client. The client can send natural language questions like asking which coins have positive 24h changes or which coins dominate the trend by market cap. The server provides two tools: one to fetch a Markdown document of CoinGecko’s trending page and another to guide parsing that document into structured data for analysis.

How to install

Prerequisites you need before start:

Python 3.10+

Node.js (required for Playwright browser dependencies)

Claude Desktop (optional, for testing with an MCP client)

Concrete steps to set up the server locally:

Clone the repository and set up the project locally.

Configuration and run instructions

Configure the MCP client to connect to your local server. The following example shows how to run the server using uv from the local directory. Replace the absolute path with the location where you placed the MCP source.

{
  "mcpServers": {
    "crypto_trending": {
      "command": "uv",
      "args": [
        "--directory",
        "/absolute/path/to/crypto-trending-mcp",
        "run",
        "main.py"
      ]
    }
  }
}

Notes on running and testing

After configuring the client, restart the MCP interface to load the new server. You should see an MCP tool indicator in the input area, allowing you to invoke the get_trending_md_doc tool and parse_trending_md_doc prompt to analyze trending data.

Examples and expected outputs

You can ask questions like: "Show the top 3 trending coins by market cap" or "Which coins have positive 24h price changes?" The server will fetch the Markdown document of the trending page, parse it, and return structured data you can use for further analysis.

Security and best practices

Keep your local environment secure and restrict access to the MCP client when testing. Do not expose the local server to untrusted networks without proper authentication and access controls.

Troubleshooting

If the MCP client cannot see the server, verify the directory path is correct and that the uv process has permission to run main.py. Ensure Python 3.10+ is installed and that Playwright dependencies can be installed by the client.

Available tools

get_trending_md_doc

Returns a Markdown document of CoinGecko's trending page, including a table with Rank, Name, Symbol, Price, 1h Change, 24h Change, 7d Change, 24h Volume, and Market Cap.

parse_trending_md_doc

Guides the MCP client to extract the trending coins table from the Markdown document and parse it with an LLM for structured analysis.