Vision MCP Server for AI Agents - give text-only models eyes. Supports any OpenAI-compatible vision API (GLM-4V, Qwen-VL, GPT-4o). Provides imageanalysis, imageanalysisurl, extracttext, diagnoseerror, understanddiagram, analyzechart tools. 触发词:图片分析、识图、OCR、截图识别、视觉、vision、image、screenshot、diagram、chart.
Portable AI agent & skills framework - write once, run on any IDE. Switch between Claude Code, Cursor, Windsurf, Copilot, and Aider without losing context. Runs locally from the agentport npm package.