lark-whiteboard

Query and edit Feishu whiteboards via lark-cli, exporting images and rendering diagrams from DSL, SVG, or Mermaid.

3|2|Updated Jul 21, 2026
One-click install
npx skills add https://github.com/zlsbksdxl/codex-lark --skill lark-whiteboard-zlsbksdxl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: lark-whiteboard
Source: https://github.com/zlsbksdxl/codex-lark/tree/main/plugins/feishu2codex/skills/lark-whiteboard
Command: npx skills add https://github.com/zlsbksdxl/codex-lark --skill lark-whiteboard-zlsbksdxl

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Creating and editing diagrams inside Feishu/Lark whiteboards normally requires manual drawing in the app. This Skill lets an agent programmatically read whiteboard content, export it as images or code, and write complex diagrams (architecture, flowchart, fishbone, funnel, flywheel, charts) directly into a whiteboard. ## Core Features & Use Cases - Whiteboard Query: Export a whiteboard as a preview image, SVG vector graphic, PlantUML/Mermaid code, or raw OpenAPI node JSON via lark-cli whiteboard +query. - Whiteboard Update: Write content back using Mermaid, PlantUML, SVG, or raw OpenAPI node format with idempotent tokens and overwrite control via lark-cli whiteboard +update. - Guided Diagram Authoring: A DSL-based rendering pipeline (via @larksuite/whiteboard-cli) with scene guides for architecture diagrams, flowcharts, comparison tables, fishbone, funnel, flywheel, bar/line charts, and more, including layout, connector, typography, and styling rules. - Use Case: Ask the agent to draw a microservices architecture diagram into a Feishu doc's whiteboard; it generates DSL JSON, renders a PNG preview for verification, converts to OpenAPI format, and writes it into the board. ## Quick Start Ask the agent to export the whiteboard in your Feishu document as a preview image, or to draw an architecture diagram into a specified whiteboard token.

Frequently Asked Questions about lark-whiteboard

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

FAQPage Schema
How do I export a Feishu whiteboard as an image?

Run lark-cli whiteboard +query with --whiteboard-token, --output_as image, and --output pointing to a PNG path. You need read permission on the whiteboard, and the command supports svg, code, and raw output formats as well.

How do I draw a flowchart or architecture diagram into a Feishu whiteboard?

Generate a diagram using the DSL path (or Mermaid/PlantUML for supported chart types), render a PNG preview with @larksuite/whiteboard-cli for verification, then convert to OpenAPI format and pipe it to lark-cli whiteboard +update with --input_format raw.

What input formats does whiteboard +update support?

The update command accepts raw (Feishu OpenAPI native node format), plantuml, mermaid, and svg via --input_format. Content can be piped from stdin with --source - or read from a file with --source @path.

Can I edit an existing whiteboard instead of recreating it?

Yes. First query the board with --output_as code to check for Mermaid/PlantUML source; if present, modify the code and re-update. Otherwise use the SVG edit route for visual edits, or query raw nodes for minor text and color tweaks.

Why does writing an image node to a whiteboard fail with a 500 error?

Image nodes require a whiteboard-domain media token. Upload the image first with lark-cli docs +media-upload --parent-type whiteboard --parent-node <board_token>, then reference the returned file token in the DSL image node.

What are the limitations of the Mermaid path for whiteboard diagrams?

The Mermaid path is intended for mind maps, sequence diagrams, class diagrams, pie charts, and Gantt charts. Complex diagrams like architecture, fishbone, funnel, or flywheel should use the DSL or SVG rendering routes for proper layout control.