← 返回 Skills 市场
MiniMax Vision Analysis
作者
daidai8910g
· GitHub ↗
· v1.0.0
· MIT-0
308
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install minimax-vision-analysis
功能描述
Analyze, describe, and extract information from images using the MiniMax vision MCP tool. Use when: user shares an image file path or URL (any message contai...
安全使用建议
This skill appears to be a wrapper for a paid MiniMax image-analysis tool and the SKILL.md instructs how to configure an MCP with MINIMAX_API_KEY and MINIMAX_API_HOST. Before installing: (1) confirm you actually have a MiniMax Token Plan and are willing to provide the MINIMAX_API_KEY to the agent, (2) ask the publisher to add the required env vars to the registry metadata (currently none are declared), (3) be aware the skill auto-triggers on many image-related keywords and extensions — if you don’t want automatic analysis, avoid enabling or request a less-broad trigger, (4) note that MCP configuration can grant local file access to images, so only proceed if you trust the source and understand where image data will be sent, and (5) if you cannot verify the MiniMax service URL or publisher, do not install. If you want me to, I can draft a short checklist or a safer SKILL.md change request to send to the skill author (e.g., declare required env vars, narrow triggers, and clarify what environment detection does).
功能分析
Type: OpenClaw Skill
Name: minimax-vision-analysis
Version: 1.0.0
The skill bundle provides instructions for an AI agent to perform image analysis using the MiniMax vision MCP tool. It includes legitimate setup procedures for various environments (OpenCode, Claude Code, Cursor) and references the 'minimax-coding-plan-mcp' package. The instructions and workflow are consistent with the stated purpose, and external links point to official documentation at platform.minimaxi.com.
能力评估
Purpose & Capability
The skill claims to call a MiniMax MCP tool for image analysis and the SKILL.md provides mode-specific prompts and configuration instructions consistent with that purpose. However, the runtime instructions require a MiniMax Token Plan and MINIMAX_API_KEY / MINIMAX_API_HOST, but the registry metadata lists no required environment variables or primary credential — this is an inconsistency between claimed requirements and declared metadata.
Instruction Scope
The SKILL.md instructs the agent to auto-trigger on any message with various image file extensions or many image-related keywords, which is broad and may cause unexpected or unwanted invocations. It also tells the agent to "detect the user's environment" and emit exact MCP setup commands for different platforms — the detection step is vague and could lead to reading environment/system details if implemented, but the file does not explicitly constrain what may be read.
Install Mechanism
This is an instruction-only skill with no install spec and no code files, so nothing will be written or downloaded during install. That lowers the attack surface; the SKILL.md points to an external docs URL for guidance but does not cause arbitrary code execution by itself.
Credentials
The instructions clearly require MINIMAX_API_KEY and MINIMAX_API_HOST and state a paid Token Plan is required, yet the skill registry lists no required environment variables or primary credential. Requesting a service API key for a vision tool is reasonable, but the omission in metadata is a mismatch that could hide credential requirements. The skill also mentions local file access via MCP if configured — users should be aware this enables reading local image files when set up.
Persistence & Privilege
The skill does not request permanent/always-on status, does not modify other skills' configs, and has no install-time hooks. Autonomous invocation is allowed by platform default (disable-model-invocation: false) but that is normal; nothing in the skill requests elevated persistence privileges.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install minimax-vision-analysis - 安装完成后，直接呼叫该 Skill 的名称或使用
/minimax-vision-analysis触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release of vision-analysis skill.
- Enables image analysis via the MiniMax_understand_image MCP tool for general description, OCR, UI review, chart data extraction, and object detection modes.
- Automatically triggers on image file uploads/links or relevant analysis requests.
- Includes guidance for MiniMax MCP setup across OpenCode, Cursor, and Claude Code environments.
- Provides mode-specific prompt strategies and structured output formats for clear results.
- Supports common image formats and local file paths (with appropriate MCP configuration).
元数据
常见问题
MiniMax Vision Analysis 是什么？
Analyze, describe, and extract information from images using the MiniMax vision MCP tool. Use when: user shares an image file path or URL (any message contai... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 308 次。
如何安装 MiniMax Vision Analysis？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install minimax-vision-analysis」即可一键安装，无需额外配置。
MiniMax Vision Analysis 是免费的吗？
是的，MiniMax Vision Analysis 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
MiniMax Vision Analysis 支持哪些平台？
MiniMax Vision Analysis 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 MiniMax Vision Analysis？
由 daidai8910g（@daidai8910g）开发并维护，当前版本 v1.0.0。
推荐 Skills