Outsource MCP

Outsource MCP is a unified MCP server that lets AI applications outsource tasks to many model providers through a single, consistent interface. It enables you to generate text and images from multiple providers without changing your client code, simplifying integration and expanding capabilities.

How to use

You interact with the Outsource MCP server from your MCP client to generate text or images through a wide range of providers. Use the same tool names and prompts you would for a single-provider setup, but select the provider you want and the model you want to run. Combine providers in your workflow to compare outputs, switch models on the fly, and build richer AI experiences for your applications.

How to install

Prerequisites: you need Python 3.11 or higher and the uvx package manager.

Step 1: Clone the MCP server repository

git clone https://github.com/gwbischof/outsource-mcp.git
cd outsource-mcp

Step 2: Synchronize dependencies with uvx

uv sync

Optional: Run the MCP Inspector to test interactively

mcp dev server.py

Step 3: Run tests (requires API keys to be set in your environment)

uv run pytest

Configuration and usage details

Configure the MCP client with a stdio server entry that points to the Outsource MCP runner and its environment. The following configuration shows the required runtime and environment variables for integrating with multiple providers.

{
  "mcpServers": {
    "outsource_mcp": {
      "command": "uvx",
      "args": ["--from", "git+https://github.com/gwbischof/outsource-mcp.git", "outsource-mcp"],
      "env": {
        "OPENAI_API_KEY": "your-openai-key",
        "ANTHROPIC_API_KEY": "your-anthropic-key",
        "GOOGLE_API_KEY": "your-google-key",
        "GROQ_API_KEY": "your-groq-key",
        "DEEPSEEK_API_KEY": "your-deepseek-key",
        "XAI_API_KEY": "your-xai-key",
        "PERPLEXITY_API_KEY": "your-perplexity-key",
        "COHERE_API_KEY": "your-cohere-key",
        "FIREWORKS_API_KEY": "your-fireworks-key",
        "HUGGINGFACE_API_KEY": "your-huggingface-key",
        "MISTRAL_API_KEY": "your-mistral-key",
        "NVIDIA_API_KEY": "your-nvidia-key",
        "OLLAMA_HOST": "http://localhost:11434",
        "OPENROUTER_API_KEY": "your-openrouter-key",
        "TOGETHER_API_KEY": "your-together-key",
        "CEREBRAS_API_KEY": "your-cerebras-key",
        "DEEPINFRA_API_KEY": "your-deepinfra-key",
        "SAMBANOVA_API_KEY": "your-sambanova-key"
      }
    }
  }
}
```"],

Notes and provider details

You can use a variety of providers for text generation and, for OpenAI, image generation with DALL-E models. Only configure API keys for providers you plan to use.

Troubleshooting

If you encounter issues, verify that the provider names match the supported list and that your API keys are correctly set in the environment. Restart your MCP client after updating configuration.

Available tools

outsource_text

Creates an Agno agent with a specified provider and model to generate text responses.

outsource_image

Generates images using AI models and returns the URL of the generated image.