giphy-mcp 2.0.0

A Model Context Protocol (MCP) server that empowers AI agents to search and retrieve lively GIFs from Giphy, enhancing chat and documentation experiences.

Giphy MCP Server

An MCP (Model Context Protocol) server that allows AI agents to search and retrieve GIFs using the Giphy API. This provides a rich way for agents to express humor, reactions, or find relevant visual context.

Prerequisites

• Python 3.10+
• A Giphy API Key. You can get a free API key at the Giphy Developer Portal.

Installation

For general use, install the package via pip:

pip install giphy-mcp

For development:

1. Clone the repository:

   git clone https://github.com/Szatek/giphy-mcp-server.git
   cd giphy-mcp-server
   

2. Create a virtual environment and install in editable mode:

   python -m venv venv
   # On Windows
   .\venv\Scripts\activate
   # On Linux/macOS
   source venv/bin/activate
   
   pip install -e ".[dev]"
   

Usage

The Giphy MCP Server is designed to be run as a standard IO transport process, which AI agents and MCP clients can spawn.

To start the server:

giphy-mcp run

To see the help menu and configuration options:

giphy-mcp help

Configuration

The server requires a Giphy API Key to function. It reads this from the GIPHY_API_KEY environment variable.

You can configure this by:

1. Creating a .env file in the project root:

   GIPHY_API_KEY=your_api_key_here
   

2. Or by passing it directly via your MCP client's configuration.

Available Tools

• search_gifs
  • Description: Search for GIFs using the Giphy API.
  • Parameters:
    • query (string): The search term to find GIFs for.
    • limit (int, optional): The maximum number of GIFs to return (default: 5).
    • rating (string, optional): Filters results by specified rating (default: 'g').

Example Use Case (Agent Interaction)

When this server is registered with an AI agent (like Claude Desktop or Cursor), the agent can autonomously use it to attach GIFs to its responses.

User: "Can you celebrate that we finally deployed to production?"

Agent (Internal Thought): The user wants to celebrate. A celebratory GIF would be perfect here. I will use the search_gifs tool with the query "celebrate".

Agent (Tool Call):

{
  "name": "search_gifs",
  "arguments": {
    "query": "celebrate",
    "limit": 1
  }
}

Agent (Response): "Woohoo! That's awesome news! Here's a celebratory GIF to match the mood:

"

Antigravity AI Agent Integration & Skills

This repository includes custom Agent Skills located in the .agents/skills directory designed for the Antigravity AI framework (or other compatible systems).

Included Skills

• Giphy Chat Enhancer: Instructs your AI agent to embed an appropriate GIF during special conversational events (successes, errors, debugging milestones), injecting a fun, nerdy communication style into your interactions.
• Giphy Documentation Enhancer: Instructs the agent to proactively embed tasteful Giphy images when generating or updating README.md files and user documentation.

Auto-Installation

The skills are written with internal logic that instructs the AI agent to recognize if it lacks the search_gifs tool. If the tool is missing, the agent will autonomously:

1. Run pip install giphy-mcp
2. Configure itself securely by modifying its $HOME/.gemini/antigravity/mcp_config.json configuration file, adding the following snippet:

"mcpServers": {
  "giphy": {
    "command": "giphy-mcp",
    "args": ["run"],
    "env": {
      "GIPHY_API_KEY": "your_api_key_here"
    }
  }
}

Using with built-in MCP Clients

To use this with an MCP client like Claude Desktop, add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "giphy": {
      "command": "giphy-mcp",
      "args": ["run"],
      "env": {
        "GIPHY_API_KEY": "your_api_key_here"
      }
    }
  }
}

Note: You may need to specify the absolute path to the giphy-mcp executable if it's not in your system's PATH.