minimax-image-understand

MiniMax MCP Image Understanding

Important note: This skill is for the OpenClaw platform. If you are using Claude Code, please ignore this skill.

Image understanding via MiniMax MCP.

Trigger conditions

Use when the user sends an image and requests analysis, description, or recognition.

Dependencies

System dependencies

Dependency	Installation method
Python 3.10+	macOS:`brew install python3<br>`Linux: `sudo apt-get install python3`

Python packages

Package name	Purpose	Install command
`mcp`	MCP client library	`pip install mcp`

Prerequisites

Install dependencies:pip install mcp
Configure environment variables: copy .env.example to .env and fill in the API Key

Usage

cd ~/.openclaw/skills/minimax-image-understand/scripts
source .env
python3 image_understand.py <图片路径或URL> [提示词]

Call in code

import sys
import os
sys.path.insert(0, "~/.openclaw/skills/minimax-image-understand/scripts")
os.environ["MINIMAX_API_KEY"] = "your-key"  # 或从 .env 加载
from image_understand import understand_image
result = await understand_image("image.jpg", "描述这张图片")

Image path

Image save location:~/.openclaw/media/inbound/, filename format:{uuid}.jpg

Hermes Agent environment notes

This skill is available in the OpenClaw environment but not in Hermes Agent, reasons:

MINIMAX_API_KEY is not injected into the Hermes Agent shell environment variables
System Python is 3.9.6, themcp package requires Python 3.10+ (can be bypassed with uv run --python 3.12 --with mcp )
Image analysis will fail in Hermes; use text descriptions or OpenClaw assistance for image reading instead

Configuration file

.env file has been placed in the scripts/ directory, format:

MINIMAX_API_KEY=your-api-key