gopher

This MCP server exposes the Gopher AI-powered Twitter search capability via the MCP interface. It implements a search_twitter tool that accepts a natural language query and returns a list of matching Twitter posts along with relevance scores. The server requires you to provide a valid Gopher AI API key (GOPHER_API) and can limit the number of results through MAX_RESULTS. To use it in an MCP client, configure the client to point at the server, and invoke the search_twitter tool with a query such as “artificial intelligence trends” to retrieve relevant tweets. The server supports error handling for API key issues, network problems, and processing delays, and it can poll for asynchronous results if the underlying Gopher API requires it.

Typical workflow:

• Set up your MCP client configuration to reference the Gopher MCP server (via Docker or binary, per your deployment).
• Call the search_twitter tool with a query string.
• Receive a structured response containing tweets, metadata, and a score indicating relevance.

Prerequisites:

• Go 1.24.7 or later (for compiling from source, if choosing to build locally)
• Docker (for Docker-based deployment) or a compatible runtime
• A valid Gopher AI API key

From Source:

1. Clone the repository:

git clone https://github.com/gopher-lab/gopher-mcp-server.git
cd gopher-mcp-server

2. Build the server binary:

go build -o gopher-mcp-server ./data

3. Run the server directly (optional):

./gopher-mcp-server

Using Docker:

1. Build the Docker image:

docker build -t gopher-mcp-server .

2. Run the container (provide API key via environment):

docker run -e GOPHER_API="your-gopher-api-key" gopher-mcp-server

Prerequisite notes:

• Ensure GOPHER_API is set to a valid API key with permissions to access Twitter data via Gopher AI.
• MAX_RESULTS can be configured to limit the number of returned tweets (default in code is 15).

Tips and common considerations:

• Always secure your GOPHER_API key; avoid embedding it in client configurations or logs.
• If you encounter a 30-second polling limit for asynchronous results, increase the poll timeout or verify connectivity to the Gopher API.
• When using Docker, map environment variables consistently across clients to ensure predictable behavior.
• If you update MAX_RESULTS, ensure MCP clients align with the new limit to avoid truncation or partial results.
• Verify that the MCP client configuration references the correct server name and entry point (search_twitter) with appropriate argument shapes.