Vision MCP Server
Provides image understanding capabilities using OpenAI's vision models (e.g., gpt-4o-mini) to analyze image content and return descriptions for software development contexts.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Vision MCP Serverdescribe the image at ./ui_mockup.png"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Vision MCP Server
为 MCP 客户端(Claude Code 等)提供图片理解能力,通过阿里云百炼/OpenAI/OpenRouter 等视觉模型分析图片内容,返回面向软件开发的描述。
快速开始
pip install -e .
python -m vision_mcp_serverRelated MCP server: MCP Image Validator
环境变量
必选
变量 | 说明 |
| 百炼 API Key(默认 Provider) |
Provider 切换
变量 | 默认值 | 说明 |
|
| Provider 名称: |
| 按 Provider | 覆盖 API 端点地址 |
| 按 Provider | 覆盖模型名称 |
| 按 Provider | 覆盖 API Key |
|
| 最大输出 token 数 |
各 Provider 默认值
Provider | 模型 | 地址 |
|
|
|
|
|
|
|
|
|
Tool: image_understand
image_understand(image_path: str, prompt: str | None = None, mode: str = "quick") -> dict参数
参数 | 类型 | 默认 | 说明 |
| string | 必填 | 本地图片路径(PNG/JPG/GIF/WebP)或 HTTP URL |
| string |
| 自定义提问,不传则自动选择提示词 |
| string |
|
|
返回
{
"description": "图片内容描述(Markdown 格式)",
"model": "qwen-vl-max",
"status": "success"
}两种模式
模式 | 耗时 | 输出 | 适用场景 |
| 5-10s | 3-4 要点 | 日常识图、快速了解 |
| 15-30s | 七维度分析 | UI 还原、设计评审、图表提取 |
detailed 模式的七个分析维度
UI 布局 — 整体结构、区块位置比例
组件结构 — 按钮/表单/表格的层次嵌套
页面层级 — 信息层级关系
配色风格 — 主色调、设计风格、明暗模式
OCR 文字 — 所有可见文字及位置
图表信息 — 图表类型、数据维度、关键数值
前端实现特征 — CSS 框架、响应式、动画、图标库
Claude Code 配置
项目根目录创建 .mcp.json:
{
"mcpServers": {
"vision": {
"command": "python",
"args": ["-m", "vision_mcp_server"],
"cwd": "E:/MCP",
"env": {
"DASHSCOPE_API_KEY": "sk-xxx"
}
}
}
}安装后 /mcp → Reconnect 生效。
项目结构
src/vision_mcp_server/
├── __init__.py
├── __main__.py # 入口
├── server.py # FastMCP + image_understand tool
├── vision.py # 多 Provider 视觉客户端
└── image_utils.py # 图片路径检测 + Base64 编码This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/GuanJiu173/DeepSeek-Vision-MCP'
If you have feedback or need assistance with the MCP directory API, please join our Discord server