Scenario

You can access Scenario.com’s generative AI API through an MCP server, enabling you to generate images from text or references, train and manage models, upscale and edit images, and access 70+ other tools directly from your MCP client. This server helps you integrate powerful AI capabilities into your workflows with simple, consistent commands.

How to use

Connect to the Scenario MCP server from your MCP client to perform a wide range of AI-powered tasks. Use the available tools to generate media, manage assets and models, and train new models. You can configure credentials once per session or provide persistent API keys for automated setups, and you can run commands locally through your MCP client without managing separate API calls.

How to install

Prerequisites: ensure you have Node.js and npm installed on your system. You will run the MCP server via npx from your MCP client configuration.

Step 1: Add the MCP server configuration to your MCP client configuration. Use the exact snippet below to enable the Scenario MCP server with required credentials.

{
  "mcpServers": {
    "scenario": {
      "command": "npx",
      "args": ["-y", "scenario.com-mcp-server"],
      "env": {
        "SCENARIO_API_KEY": "your_api_key_here",
        "SCENARIO_API_SECRET": "your_api_secret_here"
      }
    }
  }
}

Alternative: Session-based Credentials

If you prefer not to use environment variables, you can configure credentials per session. The MCP client will prompt you to set them when needed.

{
  "mcpServers": {
    "scenario": {
      "command": "npx",
      "args": ["-y", "scenario.com-mcp-server"]
    }
  }
}

Then in your MCP client:

Use set-credentials with API key api_xxx and secret yyy

## Available tools

### set-credentials

Configure API access for a session by setting the API key and secret.

### get-credentials-status

Check the current credential status to verify access is active.

### post-txt2img-inferences

Generate images from text prompts using the text-to-image model.

### post-img2img-inferences

Generate images from a source image with optional prompts for style or alterations.

### post-controlnet-inferences

Generate images using ControlNet guidance for finer control over the output.

### post-inpaint-inferences

Edit or repair specific regions of an image through inpainting.

### post-upscale-inferences

Enhance image resolution while preserving or improving quality.

### post-remove-background-inferences

Remove the background from an image.

### get-assets

List all assets stored in the Scenario MCP workspace.

### delete-asset

Remove an asset from the workspace by identifier.

### get-models

List available AI models or trained custom models.

### post-models

Create a new custom model with specified parameters and metadata.

### put-models-train-by-model-id

Initiate training for a specific model identified by its ID.

### post-models-training-images-by-model-id

Add training images to a specific model during the training process.