VisionPower

RunhuaHuang/VisionPower
★ 54 stars JavaScript AI/LLM Updated 3mo ago ⚠ Review
VisionPower: portable image understanding MCP server for Codex, Claude Desktop, Cursor, and other agents.
View on GitHub → 🛡️ Security & TRiSM Audit →

⚡ What This Server Does & Key Capabilities

The VisionPower Model Context Protocol (MCP) server provides seamless integration between AI assistants (such as Claude Desktop, Cursor, and Claude Code) and VisionPower: portable image understanding MCP server for Codex, Claude Desktop, Cursor, and other agents. By connecting this server, your AI agent gains specialized capabilities to inspect, query, and automate tasks within the AI/LLM domain.

Core Agent Capabilities

💡 Example Prompts to Ask Your AI Agent

Once installed in Claude Desktop or Cursor, try asking your assistant:

"Route this complex analysis task to the connected model and return a structured JSON report."
"Compare the reasoning outputs between two connected models and highlight any discrepancies."

Compatible AI Clients

🟣 Claude Desktop ⚡ Cursor IDE 💻 Claude Code CLI 🌊 Windsurf 🤖 Gemini CLI

Quick Install

Copy the config for your editor. Verify environment variables or setup notes in the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "visionpower": {
      "command": "npx",
      "args": [
        "-y",
        "RunhuaHuang/VisionPower"
      ]
    }
  }
}
⚡
Building Production MCP Agents? PRO TOOLKITS
Get the Complete 31,400+ MCP Ecosystem Dataset ($29) or the FastMCP + TypeScript Starter Kit ($39) with turnkey Docker & Cloudflare deployment.

README Documentation

<div align="center"> 给你的 AI Agent 装上眼睛 —— 一个轻量、安全、即插即用的图片理解能力,同时支持 MCP 与 Skill 两种接入形态。 </div> VisionPower 让 Codex、Claude Desktop、Cursor、Cline、Cherry Studio 等 Agent 获得识别图片内容、读取截图文字(OCR)、解读图表、按顺序分析多张图片的能力。 它不绑定任何模型:默认走阿里云百炼 / DashScope 的 Qwen-VL(OpenAI-compatible 接口),也可通过模型名和 Base URL 配置切换到智谱 GLM、MiniMax、Kimi、火山方舟豆包、Google Gemini、GPT-4o 或任何兼容 OpenAI /chat/completions 视觉输入的服务。同一套内核提供两种接入形态——MCP 和 Skill,按你的 Agent 能力任选其一或都装。

View Full Documentation & Updates on GitHub ↗

Tools (20)

VISIONPOWER_ALLOWED_DIRSVISIONPOWER_API_KEYVISIONPOWER_BASE_URLVISIONPOWER_CACHEVISIONPOWER_CACHE_MAX_ENTRIESVISIONPOWER_CACHE_TTL_MSVISIONPOWER_DEBUGVISIONPOWER_INBOX_DIRVISIONPOWER_INBOX_MAX_ENTRIESVISIONPOWER_INBOX_TTL_MSVISIONPOWER_MAX_IMAGESVISIONPOWER_MAX_IMAGE_BYTESVISIONPOWER_MAX_RETRIESVISIONPOWER_MAX_TOKENSVISIONPOWER_MAX_TOTAL_IMAGE_BYTESVISIONPOWER_MODELVISIONPOWER_SKILL_STATEVISIONPOWER_TIMEOUT_MSimage_base64image_mime_type
📊
Building AI Agents or Backtesting Quant Models?
Get the Polymarket Historical Dataset — 23.0M+ 15-minute price & orderbook snapshots across 18,400+ prediction markets. Clean CSV & SQLite download for backtesting & agent training.
Buy Dataset ($19) → View Sample →