A technical utility for programmatically extracting image assets and vector board exports from Feishu (Lark) Whiteboards.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install feishu-whiteboard-extract
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install feishu-whiteboard-extract using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Feishu Whiteboard Extract?
Feishu Whiteboard Extract is a specialized tool within the Openclaw Skills library designed to handle the extraction of visual data from Feishu Whiteboards. It primarily functions by identifying all image nodes within a specified board and retrieving their corresponding file tokens. This allows for automated downstream processes such as local archiving, OCR processing, or migrating visual assets to other platforms.
In addition to individual node extraction, the skill supports a comprehensive full-board export feature. This functionality generates high-fidelity vector representations (SVG) or fallback raster images of the entire board layout. By integrating this skill into an AI agent workflow, developers can ensure that visual brainstorming sessions are captured, structured, and made searchable within a wider knowledge management system.
Feishu Whiteboard Extract Use Cases
- Automated OCR pipelines for converting whiteboard diagrams and handwritten notes into structured text.
- Archiving collaborative brainstorming sessions into a corporate knowledge base in SVG format for high-resolution review.
- Processing embedded whiteboards within Feishu Docx files by identifying and extracting tokens from specific block types.
- Bulk downloading of visual assets for audit logs or technical documentation attachments.
How Feishu Whiteboard Extract Works
- The user provides a whiteboard_id, typically extracted from a Feishu URL or a document block.
- The extract_images.js script executes a list request to the Feishu board.v1.whiteboardNode.list API.
- The skill parses the response to isolate image nodes, generating a JSON map of node_id and file_token pairs.
- For full-board context, the export_board_svg.js script invokes the download_as_image OpenAPI endpoint with an SVG preference.
- The resulting file tokens or image data are then ready for processing by other Openclaw Skills, such as media downloaders or vision-based AI models.
Feishu Whiteboard Extract Setup
To get started with this skill, ensure your bot has 'View' permissions on the target whiteboard. Run the following command to extract image tokens:
node skills/feishu-whiteboard-extract/extract_images.js <whiteboard_id>
To export the entire board as a vector graphic for your archives:
node skills/feishu-whiteboard-extract/export_board_svg.js <whiteboard_id> appendix/board_full.svg
Feishu Whiteboard Extract Data Schema & Taxonomy
The extraction process generates a structured JSON output with the following schema:
| Attribute | Type | Description |
|---|---|---|
| whiteboard_id | String | The unique identifier of the source board. |
| count | Integer | Number of image nodes identified. |
| images | Array | A list containing the node_id and the file_token for each image. |
| file_token | String | The media-specific token used for secure image downloading via the drive API. |
Feishu Whiteboard Extract Advanced Features
- Smart Endpoint Routing: Automatically uses the drive/v1/medias endpoint instead of the standard files endpoint to ensure successful downloads without 403 errors.
- Vector-First Export: Prioritizes image/svg+xml formats to maintain maximum visual fidelity during board exports.
- Docx Block Interoperability: Specifically designed to handle block_type 43, allowing agents to find whiteboards embedded within standard Feishu documents.
- Multi-format Fallback: Intelligently handles content-type responses to save files with correct extensions (PNG, JPG, or SVG) based on server availability.
SKILL.md
Loading
Related Openclaw Skills
Baidu Netdisk MCP Skill
Manage and automate Baidu Netdisk file operations directly through Openclaw Skills using a secure, hot-reloading MCP server.
AKShare Data Router (CN)
A sophisticated data routing skill that maps Chinese financial market queries to specific AKShare data sources and analytical processing recipes.
Project Code Standard
A comprehensive automation skill for checking, fixing, and enforcing code standards across Python, JavaScript, and TypeScript projects.
Subagent-Driven Development
A sophisticated workflow that dispatches fresh subagents per task to ensure high-quality, parallel-safe execution within a single development session.
Market Chart Renderer
A high-performance rendering engine that converts standardized market data into interactive HTML and static PNG charts with technical indicators.
Apple Music Control
An AI-driven skill for controlling Apple Music playback, searching library catalogs, and managing playlists directly from the command line on macOS.
Featured*
