AgentSkillsCN

Visionary Skill - Visual Architecture & Diagramming

远见技能 - 可视化架构与图表绘制

SKILL.md

Visionary Skill - Visual Architecture & Diagramming

namedescription
visionaryTransforms text-based descriptions into visual diagrams and architectural sketches. USE WHEN you need to visualize a system, workflow, or data model.

The Key Insight

A picture is worth a thousand lines of code. Visionary allows the PAI to "see" the architecture it is building. By generating Mermaid or Excalidraw definitions, it provides a visual feedback loop for both the user and the agent.

Usage

Generate Sequence Diagram

bash
pai run Visionary mermaid "Alice -> Bob: Hello"

Sketch Architecture

bash
pai run Visionary sketch "3-tier web app with redis cache"

Data Model Diagram

bash
pai run Visionary db "User table, Post table with foreign key"

How it Works

It utilizes Mermaid.js syntax templates. When a request is made, it fills a template with the provided logic and outputs the raw code (which many CLIs/IDEs can render natively) or saves it as a .mmd file.

Strategic Value

  • Complex System Design: Makes it easier to track state transitions in distributed systems.
  • Documentation Excellence: Automatically generates architecture diagrams for your README.md or History/Decisions/.
  • Stakeholder Communication: Provides a high-level view for non-technical users.