Program120

Generate interactive, draggable diagrams with REAL-TIME incremental rendering via SSE. Supports flowcharts, architecture diagrams, mind maps, and sequence diagrams. Nodes appear one by one as the agent generates them — zero wait time. Use this skill whenever the user asks to draw, visualize, or create any kind of diagram. Trigger keywords: 画图, 流程图, 架构图, 思维导图, 时序图, diagram, flowchart, architecture, mind map, sequence diagram, 画一个, 可视化, visualize, 拓扑图, 关系图, draw, 图解, 示意图. Also trigger when users describe processes, architectures, or relationships that would benefit from visual representation, even if they don't explicitly say "draw".

Generate AI images locally by delegating to the installed Codex CLI's GPT-native image_gen tool. Use this skill whenever the user asks to generate, create, draw, render, or make any kind of bitmap image, picture, illustration, photo, hero, banner, poster, sprite, mockup, icon-as-bitmap, or visual asset — including 生图, 画一张, AI 生图, AI 画图, make me a picture, draw me, render an illustration — and Claude does not otherwise have a working native image generation tool in this session. Also use it when the deliverable for a task is a PNG/JPG (UI mockup, blog hero, social card, product shot, concept art, game asset). Do NOT use for matching an existing SVG/vector icon system, simple shapes/diagrams that should stay as code (HTML/CSS/canvas/SVG), or for editing a user-supplied image where they only asked for trivial cropping/resizing — defer to the user's tools or simpler approaches in those cases.