Get the FREE Ultimate OpenClaw Setup Guide →

VisionCraft

VisionCraft MCP delivers up-to-date, specialized computer vision and Gen-AI knowledge directly to Claude and other AI assistants.

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio augmentedstartups-visioncraft-mcp-server npx -y visioncraft-mcp@latest --api-key YOUR_API_KEY_HERE \
  --env API_KEY="YOUR_API_KEY_HERE"

How to use

VisionCraft is an MCP server that provides specialized computer vision and GenAI knowledge through the Raven engine and a dedicated toolset. The primary accessible tool is vision-query, which lets you retrieve up-to-date information from VisionCraft's knowledge base, including topics like object detection, segmentation, 3D vision, and vision-language models. To use VisionCraft in your prompts, reference the tool as visioncraft mcp (for example, add a directive like use visioncraft mcp in Claude or your MCP client) so the assistant can route queries to the knowledge base and return context-aware results. The server is designed to stay current with the latest models and techniques, and it supports integration with common MCP clients such as Claude Desktop, Windsurf, Cursor, and VS Code/Cline.

How to install

Prerequisites:

  • Node.js v18 or newer
  • Access to an MCP-compatible client (e.g., Claude Desktop, Windsurf, Cursor, VS Code/Cline)

Installation steps:

  1. Install Node.js if you haven’t already. Visit https://nodejs.org/ and install the LTS release for your platform.
  2. Ensure you have an API key for VisionCraft. Sign up at visioncraft-app.augmentedstartups.com and obtain your API key.
  3. Install VisionCraft MCP via npx by using the configuration example below in your MCP client configuration file.

Example configuration (for Claude Desktop, Windsurf, Windsurf-like clients, or VS Code Cline):

{
  "mcpServers": {
    "visioncraft": {
      "command": "npx",
      "args": ["-y", "visioncraft-mcp@latest", "--api-key", "YOUR_API_KEY_HERE"]
    }
  }
}
  1. Replace YOUR_API_KEY_HERE with your actual VisionCraft API key. Save the configuration file in the appropriate location for your MCP client (e.g., claude_desktop_config.json for Claude Desktop).
  2. Start your MCP client and enable the VisionCraft server as configured. You should now be able to issue prompts that leverage VisionCraft’s vision and GenAI knowledge via the vision-query tool.

Additional notes

Notes and tips:

  • You can pass the API key in the config or rely on the env variable API_KEY if your client supports reading environment variables. Keep your API key secure and avoid committing it to public repos.
  • The mcpServers key can be named differently in some clients (e.g., servers). If your client requires a different top-level key, adjust accordingly.
  • If you encounter authentication errors, double-check that the API key is valid and that the client has network access to VisionCraft services.
  • The vision-query tool is the primary means of retrieving knowledge; you can extend prompts with explicit references to visioncraft mcp to ensure routing to the correct knowledge base.
  • Ensure Node.js version compatibility (v18+). Some MCP clients may have their own version requirements.
  • If you’re using multiple MCP servers, ensure each server has a unique key under mcpServers to avoid conflicts.

Related MCP Servers

Sponsor this space

Reach thousands of developers