MiniMax MCP for Openclaw

A dual-capability skill providing AI agents with real-time internet search and sophisticated multi-modal image analysis.

hongjiahao371-pixel
v1.0.0
Mar 22, 2026
0
726
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install minimax-mcp-tools

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install minimax-mcp-tools using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is MiniMax MCP?

The MiniMax MCP is an official Model Context Protocol service designed to bridge the gap between static LLM knowledge and the real-time web. By integrating this skill, AI agents gain the ability to perform live web searches and interpret visual data from both URLs and local file paths. It serves as a critical utility for developers building agents that require up-to-date information and vision-based reasoning.

Built to work seamlessly within the Openclaw Skills ecosystem, this tool leverages the MiniMax platform's robust API to deliver high-accuracy search results and detailed image descriptions. Whether you are debugging visual UI elements or researching the latest tech trends, this skill provides the necessary eyes and ears for your autonomous agents.

MiniMax MCP Use Cases

  • Real-time research and fact-checking during AI development sessions.
  • Automated visual analysis of screenshots or UI mockups for frontend verification.
  • Gathering current market data or news to inform agent-driven decision making.
  • Processing local image files for OCR or descriptive metadata generation.

How MiniMax MCP Works

  1. The user or agent triggers a request for information or image analysis.
  2. The Openclaw gateway identifies the need for the MiniMax MCP tool based on the intent.
  3. For searches, the skill queries the MiniMax web search engine and returns structured snippets.
  4. For images, the skill processes the provided URL or local path through the MiniMax vision model.
  5. The refined context is injected back into the AI's prompt for a final, informed response.

MiniMax MCP Setup

First, ensure you have uvx installed on your system. Then, install the skill via the CLI:

clawhub install minimax-mcp

Configure your ~/.openclaw/mcp.json by adding your MiniMax API Key:

{
  "mcpServers": {
    "MiniMax": {
      "command": "uvx",
      "args": ["minimax-coding-plan-mcp", "-y"],
      "env": {
        "MINIMAX_API_KEY": "YOUR_API_KEY",
        "MINIMAX_API_HOST": "https://api.minimaxi.com"
      }
    }
  }
}

Restart the gateway to apply changes:

openclaw gateway restart

MiniMax MCP Data Schema & Taxonomy

The skill utilizes two primary tools with specific input schemas:

Tool Parameter Description
web_search query The string keywords used for the live search engine request.
understand_image prompt, image_url A descriptive instruction and the source (URL or local path) of the image.

Supported image formats include JPEG, PNG, GIF, and WebP with a maximum file size of 20MB.

MiniMax MCP Advanced Features

  • Direct integration with the mcporter CLI for manual tool testing and debugging.
  • Multi-modal support for local file system access, allowing agents to 'see' files in the workspace.
  • Seamless compatibility with the Openclaw Skills gateway for automated tool selection during natural language dialogues.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*