Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Updated documentation to full English, replacing the previous Chinese SKILL.md with an English version. - Improved clarity and consistency in setup instructions, usage scenarios, and environment variable descriptions. - Adjusted installation and configuration guidance for internationalization. - No backend or functional changes; update is documentation only.
What This Skill Does
MCP server for MiniMax's AI API with web search and image analysis capabilities. Supports both global (minimax.io) and China (minimaxi.com) API endpoints. Exposes two tools: web_search for structured results and understand_image for AI-powered image recognition.
Combines web search and image understanding in a single MCP server, with dedicated regional endpoints for users in mainland China.
When to use it
- Searching the web for recent news or events
- Describing what's shown in a photo or screenshot
- Extracting readable content from a webpage URL
- Identifying objects or text in product images
- Answering questions about image content
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: "What's in this image? [image URL]"
- 1Receives image URL and description prompt from user
- 2Calls understand_image tool with the URL and prompt
- 3MiniMax API processes the image and returns analysis
- 4Returns structured text description of image contents
Text description of the image's contents, objects, and context
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.