mcp-dexscreener is a foundational MCP (Model Context Protocol) server for AI agents that wraps Dexscreener's free open API, enabling AI agents to query real-time market prices of tokens on any chain. The project is built based on the Dexscreener API documentation (as of April 4, 2025), and the corresponding API reference documentation is https://docs.dexscreener.com/api/reference。
In terms of main capabilities, the material clearly states that it allows AI agents to use Dexscreener's free open API to check prices on any chain. The repository does not provide a more detailed feature list, so other specific capabilities are not currently provided.
In terms of usage, all dependencies first need to be installed. If using Claude Desktop, after pulling the code, open the configuration file claude_desktop_config.json in VSCode: on macOS run code ~/Library/Application\ Support/Claude/claude_desktop_config.json; on Windows run code $env:AppData\Claude\claude_desktop_config.json. For more information, see https://modelcontextprotocol.io/quickstart/server。然后在 Place the dexscreener object in claude_desktop_config.json.
When running the application, you can use the Inspector to test the MCP server without using Claude Desktop, supporting both the default STDIO version and the SSE version index-sse.js (server-sent events, which can be hosted on a remote server).
In terms of applicable scenarios, the material shows that it is suitable for scenarios where AI agents need to query real-time on-chain prices, especially when used with Claude Desktop or tested through the Inspector.
In terms of limitations, the material does not provide specific explanations regarding API rate limits, supported chain scope, fees, data accuracy, or the range of available tokens; this information is not currently provided.
