mcp -giphy

This MCP server provides an integration with the Giphy API, exposing tools to search for GIFs, fetch a random GIF, and retrieve trending GIFs. It returns structured GIF objects with metadata suitable for consumption by AI models or other MCP clients. The available tools are: search_gifs, get_random_gif, and get_trending_gifs. To use the server, configure your MCP client or Claude Desktop to point at the Giphy MCP server and supply your GIPHY_API_KEY via the environment variable. The response objects include essential fields such as id, title, url, and an images object with various formats, enabling downstream processing and attribution.

Prerequisites:

• Node.js (LTS) and npm installed on your system
• Basic familiarity with running MCP servers or using npx to start a server

Installation steps:

1. Ensure Node.js and npm are installed. Verify with: node -v npm -v

2. Install or run the MCP server using npx (as demonstrated in the README):

# Start the MCP server via npx (no local install required)
npx -y mcp-server-giphy

3. Alternatively, install locally in a project:

# Initialize a project (if needed)
npm init -y
# Install the MCP server package
npm install mcp-server-giphy

4. If you prefer a global install, you can install globally and run from anywhere:

npm install -g mcp-server-giphy
mcp-server-giphy

5. Run with environment variables (recommended):

GIPHY_API_KEY=your_api_key_here npx -y mcp-server-giphy

Note: The GIPHY_API_KEY is required for actual API access. Do not commit this key in code or public repositories.

Tips and notes:

• GIPHY_API_KEY must be provided in the environment where the server runs. Without it, GIF results will not be fetched from the Giphy API.
• The server supports content filtering via rating (g, pg, pg-13, r). Use these values when calling the tools to ensure appropriate content.
• If you are integrating with Claude Desktop or another MCP client, configure the mcpServers entry to include the env variable as shown in the README example.
• The response format includes a rich images object with various formats, helpful for selecting an appropriate image size for display or model ingestion.
• Be mindful of Giphy rate limits and potential API key restrictions based on your plan. Cache results as needed to optimize performance.
• Development and testing can be done with npm scripts if you clone or install locally; run npm run dev or npm start as configured by the package maintainers if available.