Gaia MCP Server

by SipherAGI

166 downloads Not rated yet
GitHub

About

# Gaia MCP Server > **⚠️ DEPRECATED** > > This repository is deprecated and is no longer actively maintained. Please use the new Go implementation instead: > > **👉 [github.com/SipherAGI/gaia-mcp-go](https://github.com/SipherAGI/gaia-mcp-go)** > > The Go implementation provides better performance, easier installation…

Explore

- Supports stdio and SSE communication methods.
- Integrates with Claude Desktop for in-conversation image generation.
- Provides six image tools: upload-image, create-style, generate-image, remix, face-enhancer, upscaler.
- Runs without installation via npx.
- Includes health check endpoint at /health.
- Deprecated in favor of a Go implementation.

Setting up with Highlight

This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:

  1. Download and install Highlight from highlightai.com/download
  2. Navigate to the plugins tab and select "Add Custom Plugin"
  3. Configure the plugin with the settings below
    Plugin Name Gaia MCP Server
    Command (node, npx, python, etc.)

    Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.

  4. Enable "Start Automatically" if you want the plugin to start when Highlight launches

From the repository

You can use Gaia MCP Server in multiple ways:

Run the server directly using npx without installing:


Install the package globally to run from anywhere:

bash
npm install -g @ather-mcp/gaia-mcp-server

Then run it from any terminal:

bash

gaia-mcp-server stdio --api-key=your-api-key


You can also specify a custom API URL:

bash
npx @ather-mcp/gaia-mcp-server stdio --api-key=your-api-key --api-url=https://your-custom-api-url

With the stdio method, communication with the server happens through stdin/stdout, following the MCP protocol format.

Example of sending a message through stdio:

json
{ "type": "message", "data": { "content": "Your message content here" } }
```

The Gaia MCP Server provides several AI image generation and manipulation tools that can be called by LLMs:

- upload-image: Upload images to the Gaia platform from URLs
- create-style: Create a new style in the Gaia platform using provided images
- generate-image: Generate images with Protogaia based on text prompts
- remix: Create new variations of an existing image
- face-enhancer: Enhance face details in an existing image
- upscaler: Enhance the resolution quality of image

Each tool is registered with the MCP server and can be invoked according to the Model Context Protocol standard.

Claude Desktop / Cursor

Paste into your MCP client config file to install this server.

{
    "mcpServers": {
        "gaia mcp server": {
            "gaia-mcp-server": {
                "command": "npx",
                "args": [
                    "@ather-mcp/gaia-mcp-server",
                    "sse"
                ]
            }
        }
    }
}

McpServers

{
    "gaia-mcp-server": {
        "command": "npx",
        "args": [
            "@ather-mcp/gaia-mcp-server",
            "sse"
        ]
    }
}

> ⚠️ DEPRECATED
>
> This repository is deprecated and is no longer actively maintained. Please use the new Go implementation instead:
>
> 👉 github.com/SipherAGI/gaia-mcp-go
>
> The Go implementation provides better performance, easier installation, and active maintenance.

An MCP (Model Context Protocol) server implementation for ProtoGaia, supporting both stdio and SSE (Server-Sent Events) communication methods.

Table of Contents

- Introduction
- Installation
- Usage
- Integrate with Claude Desktop
- Using Stdio Method
- Using SSE Method
- Understanding MCP
- API Endpoints
- Supported Tools
- Notice

Introduction

This package contains a TypeScript implementation of an MCP server for ProtoGaia. It provides a standardized way for LLMs (Large Language Models) to communicate with external tools and services using the Model Context Protocol.

Installation

You can use Gaia MCP Server in multiple ways:

Using with npx (No Installation)

Run the server directly using npx without installing:

```bash

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.