lark-whiteboard-cli

Generate and render diagrams on Lark whiteboards using DSL or Mermaid via whiteboard-cli.

1.3k|142|Updated Jun 1, 2026
One-click install
npx skills add https://github.com/MiniMax-AI/minimax-code --skill lark-whiteboard-cli-minimax-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: lark-whiteboard-cli
Source: https://github.com/MiniMax-AI/minimax-code/tree/main/packages/local-runtime/assets/skills/lark-tools/cli-skills/lark-whiteboard-cli
Command: npx skills add https://github.com/MiniMax-AI/minimax-code --skill lark-whiteboard-cli-minimax-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @larksuite/whiteboard-cli, lark-cli, and includes references (resource) components.

What problem does it solve? Creating architecture diagrams, flowcharts, funnels, fishbone charts, and other visualizations on Lark (Feishu) whiteboards requires precise layout, styling, and coordinate logic that is tedious to produce by hand. This Skill guides the design of whiteboard diagrams through a structured DSL or Mermaid, renders them to PNG, and optionally uploads them to a Feishu whiteboard. ## Core Features & Use Cases - Dual rendering paths: Routes mind maps, sequence diagrams, class diagrams, and pie charts to Mermaid, while all other chart types use a precise JSON DSL with Flex, Dagre, or absolute layouts. - Scene-specific guidance: Provides dedicated layout rules and script templates for architecture diagrams, flowcharts, swimlanes, comparison tables, fishbone diagrams, bar/line charts, funnels, flywheels, pyramids, treemaps, milestones, and org charts. - Render, review, and upload: Renders diagrams to PNG via @larksuite/whiteboard-cli, applies a visual self-check and symptom-to-fix table, and uploads to Feishu whiteboards through lark-cli with a mandatory dry-run overwrite check. - Use Case: Ask for a microservices architecture diagram; the Skill plans content, generates DSL JSON (or a coordinate-computing Node.js script for geometric charts), renders a PNG, and uploads it to your Feishu whiteboard after confirmation. ## Quick Start Ask the assistant to draw an architecture diagram of your system on a Feishu whiteboard and it will generate, render, and optionally upload the diagram.

Frequently Asked Questions about lark-whiteboard-cli

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I draw an architecture diagram on a Feishu whiteboard?

Describe the system you want diagrammed, and the Skill plans the content, generates a whiteboard DSL JSON file, and renders it to PNG with @larksuite/whiteboard-cli. If you provide a whiteboard token or document URL, it can also upload the result to Feishu via lark-cli.

When does the whiteboard skill use Mermaid instead of DSL?

Mind maps, sequence diagrams, class diagrams, and pie charts route to Mermaid because their structures auto-layout well. All other chart types, including flowcharts and architecture diagrams, use the JSON DSL for precise control over styling and layout.

What dependencies are required to render whiteboard diagrams?

Rendering requires Node.js 18+ and the @larksuite/whiteboard-cli npm package (version ^0.2.0), invoked via npx. Uploading to Feishu additionally requires lark-cli with proper authentication, and installing new dependencies requires user consent first.

Why does text get truncated in rendered whiteboard nodes?

Text truncation happens when a node containing text uses a fixed numeric height. The DSL requires height set to fit-content for all text-bearing nodes because the engine does not support overflow handling.

Can I update an existing Feishu whiteboard without losing content?

Yes, but the Skill enforces a mandatory dry-run check with --overwrite --dry-run before writing to an existing whiteboard token. If nodes would be deleted, it stops and asks for your confirmation, or you can request a non-overwriting update.

Why must fishbone or flywheel diagrams be generated by a script?

These charts require trigonometric or polar coordinate calculations for branch angles and ring segments, which are error-prone to handwrite. The Skill writes a Node.js script that computes coordinates and outputs the diagram JSON before rendering.