MCP.so
ログイン

Scenario AI MCP Server

@pasie15

基本情報

カテゴリ

その他

トランスポート

stdio

作者

pasie15

Scenario AI MCP Server について

A Model Context Protocol (MCP) server for the Scenario AI API

Scenario AI MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "scenario-ai-mcp-server": {
      "command": "python",
      "args": [
        "src/server.py"
      ]
    }
  }
}

Scenario AI MCP Server のツール

2

`prompt` (required): The text prompt describing the image to generate

`asset_id` (required): The asset ID of the image to process

概要

What is Scenario AI MCP Server?

The Scenario AI MCP Server is a Model Context Protocol server that integrates with the Scenario AI platform (https://app.scenario.com/). It enables developers to generate images from text prompts and remove backgrounds from images programmatically via MCP tools.

How to use Scenario AI MCP Server?

Clone the repository, install dependencies from src/requirements.txt, set up a .env file with your Scenario AI API credentials (SCENARIO_API_KEY, SCENARIO_API_SECRET, and optionally SCENARIO_MODEL_ID), and run python src/server.py or use the provided start scripts in the scripts/ folder. The server exposes MCP tools (generate_image, remove_background) and resources (status://info, job://{job_id}, asset://{asset_id}) that can be used with a compatible client like Roo-Cline.

Key features of Scenario AI MCP Server

  • Text-to-image generation using the Scenario AI API
  • Background removal from images given an asset ID
  • MCP resources for server status, job, and asset details
  • Example clients and test suite included
  • Scripts for easy startup on Windows and Linux/macOS

Use cases of Scenario AI MCP Server

  • Generating game assets or concept art from text prompts
  • Automatically removing backgrounds from product images
  • Integrating AI image generation into creative workflows via MCP
  • Building applications that require on-demand image creation and editing

よくある質問

1

What API credentials are required?

You must provide `SCENARIO_API_KEY` and `SCENARIO_API_SECRET` in a `.env` file. An optional `SCENARIO_MODEL_ID` can be set as the default model for image generation.

2

Which Python dependencies are needed?

All dependencies are listed in `src/requirements.txt` and can be installed with `pip install -r src/requirements.txt`. Additionally, the MCP SDK must be installed (`pip install mcp`).

3

Where are generated images stored?

Images are stored on the Scenario AI platform. The server provides asset IDs and job IDs to retrieve them via the API.

4

What MCP tools does the server expose?

The server exposes `generate_image` (with parameters `prompt`, `model_id`, `negative_prompt`, `num_samples`) and `remove_background` (requires `asset_id`).

5

Can I run the server without setting up a model ID?

Yes, `model_id` is optional for `generate_image`; if omitted, it defaults to the value in the environment variable `SCENARIO_MODEL_ID`. If that is also unset, the API may require a valid model ID.

コメント

「その他」の他のコンテンツ