Frequently Asked Questions about UBOS and Recraft AI MCP Server

Q: What is the Recraft AI MCP Server? A: It’s an implementation of the Model Context Protocol (MCP) that connects MCP clients with Recraft AI’s image generation and manipulation tools, allowing you to generate and edit raster and vector images using AI.

Q: What is MCP (Model Context Protocol)? A: MCP is an open protocol standardizing how applications provide context to Large Language Models (LLMs), enabling them to access external data and tools.

Q: What kind of images can I generate with Recraft AI? A: You can generate both raster and vector images from textual prompts. Also, can edit images with the prompt.

Q: What tools are available in the Recraft AI MCP Server? A: Tools include generate_image, image_to_image, create_style, vectorize_image, remove_background, replace_background, crisp_upscale, creative_upscale, and get_user for account information.

Q: Do I need a Recraft AI account to use the MCP Server? A: Yes, you need a Recraft AI API key to authenticate and use the service.

Q: How do I get a Recraft AI API key? A: Register for an account on Recraft AI and generate an API key from your profile API page.

Q: What is UBOS? A: UBOS is a full-stack AI Agent Development Platform that helps you orchestrate AI Agents, connect them with your enterprise data, build custom AI Agents, and create Multi-Agent Systems.

Q: How does the Recraft AI MCP Server integrate with UBOS? A: UBOS provides a comprehensive environment for building and deploying AI Agents, and the Recraft AI MCP Server adds a powerful visual content creation capability to this ecosystem, allowing for orchestration, data connection, customization, and multi-agent system integration.

Q: Which MCP clients are compatible with the Recraft AI MCP Server? A: Compatible MCP clients include Claude Desktop.

Q: How do I set up the Recraft AI MCP Server? A: You’ll need a Recraft AI API key, an MCP client, and need to modify the client’s configuration file (e.g., claude_desktop_config.json) to include the server’s configuration details.

Q: Can I store generated images in a specific directory? A: Yes, you can specify the IMAGE_STORAGE_DIRECTORY environment variable to store generated images in a custom location. The default directory is $HOME_DIR/.mcp-recraft-server.

Q: What are the pricing details for the Recraft AI tools? A: The price varies depending on the tool. For example, image generation costs $0.04/$0.08 per raster/vector image, and vectorization costs $0.01 per image. See Recraft AI API docs for detailed pricing.

Q: Where can I find more detailed information about the Recraft AI tools and their parameters? A: Detailed information can be found in the Recraft AI API docs.

Q: Is it possible to run the MCP Server from source code? A: Yes, you can clone the repository from GitHub, build the server using npm install and npm run build, and then configure your MCP client to run the server from the built files.