beautiful-feishu-whiteboard

$npx mdskill add JasonxzWen/harness-hub/beautiful-feishu-whiteboard

Compose editable Feishu/Lark whiteboards from native SVG shapes

  • Verify existing renderer or authenticate for live editing.
  • Check for available SVG-to-Feishu whiteboard capability.
  • Decide based on target document and authorization in request.
  • Return results as an editable Feishu/Lark whiteboard.

SKILL.md

.github/skills/beautiful-feishu-whiteboardView on GitHub ↗
---
name: beautiful-feishu-whiteboard
description: Load when creating or restyling a polished editable Feishu or Lark whiteboard, infographic, diagram, poster, or visual explainer; do not load for generic SVGs, static images, or other whiteboard platforms.
license: MIT
metadata:
  source: "zarazhangrui/beautiful-feishu-whiteboard"
  upstream_commit: "6989843b355ac92ebbd4f66166189a001e61e9b5"
  adaptation: "Preserves the verified medium rules, catalogue, and 35 palette documents while removing installers, authentication setup, automatic downloads, screenshots, and fixed CLI write commands."
---

# Beautiful Feishu Whiteboard

Compose an editable Feishu/Lark whiteboard from native SVG shapes, using one of 35 curated palette systems and the medium's verified rendering constraints.

This Skill owns the board's visual system, SVG composition, and render review. It does not authorize dependency installation, account configuration, authentication, document creation outside the request, or unrelated Feishu writes.

## Availability And Authority

Before building, perform read-only checks for:

- an already available SVG-to-Feishu whiteboard renderer or equivalent Host capability;
- an already authenticated Feishu/Lark writer if the requested outcome includes a live editable board;
- the exact existing document/whiteboard target, or clear authorization in the request to create a new board.

If a required capability is absent, report it and stop at the highest truthful artifact boundary. Do not install packages, run an auto-downloading command, configure an account, start login, or copy credentials.

A request for a local SVG or preview does not authorize a Feishu write. A request to create or update a Feishu whiteboard authorizes only the named board or the smallest new document needed for that board.

## Build

### 1. Resolve the board

Determine the content, purpose, audience, target, and whether the user needs:

- a local SVG/preview;
- an editable live whiteboard;
- both.

Ask one short question only when a missing answer changes content or the external write target.

### 2. Choose one visual system

If the user named a style, use it. Otherwise read `CATALOG.md` and choose by loudness, formality, mood, and palette.

Commit to one candidate before loading a template. Then read:

- all of `RULES.md`;
- only `templates/<chosen-slug>/design.md`.

Do not open several full palette documents to compare them. The catalogue is the comparison surface.

### 3. Compose native SVG

Use a logical canvas roughly 1600–1700 units wide and let content determine height. Choose the narrative shape that best fits the material: stages, flow, comparison, timeline, or system map.

Follow these invariants:

- use editable native rectangles, rounded rectangles, circles, ellipses, straight or right-angled connectors, and text;
- keep every label as `<text>`/`<tspan>` and do not set `font-family`;
- use native connector arrowheads via markers, never decorative triangle or chevron arrowheads;
- use only solid colors; no gradients, masks, clipping, filters, blur, or opacity;
- keep all visible copy content-bearing. Never put the prompt, source path, style name, tool name, or build notes on the board.

### 4. Render And Correct

Use an already available renderer. Inspect the rendered image, then fix:

- overflow and clipping;
- tight outer margins or internal padding;
- accidental overlaps;
- arrowhead defects;
- low contrast and tiny labels;
- content that reads in the wrong order.

Apply local SVG fixes in one targeted batch, render again, and repeat until clean. Do not regenerate the entire board to move one box.

If no compatible renderer is available, do not claim Feishu fidelity; return the SVG as unverified or stop if the user required a verified editable board.

### 5. Write Only When Authorized

After the local render passes, use the already available Feishu/Lark writer against the resolved target. Use an idempotent request mechanism when the available capability supports one.

Query the live board after the write. Inspect its rendered layout and verify stored text colors through raw/live data when image export is known to render text colors unreliably. Make only scoped corrections to the same target.

## Delivery

For a completed live board, return both the confirmed Feishu document/board link and the final rendered image. For a local-only result, return the SVG and preview paths.

State the chosen palette, validation evidence, unavailable checks, and whether the user can request a palette switch. Never claim editability, a live link, or a successful write from a local render alone.

More from JasonxzWen/harness-hub

SkillDescription
agent-interaction-auditLoad when the user asks for private agent interaction audits from local Codex/Claude Code traces, task profiles, tool-call decisions, collaboration bottlenecks, prompt/rule or automation-log audits, closeout retrospectives, or evidence-backed Skill/project-rule/Eval/SOP/OKF improvements.
agent-reachLoad when a task needs read-only internet research or content retrieval through an already installed Agent Reach CLI; diagnose available backends first, never install, upgrade, authenticate, configure, or copy credentials on the user's behalf.
animation-vocabularyLoad when a user describes a Web animation or motion effect without knowing its name and needs the precise term or close alternatives; use only for naming, not design, implementation, or review.
apple-designLoad when a frontend task needs Apple-platform design guidance or physical, gesture-driven, velocity-aware, interruptible interaction; skip generic visual styling and use review-animations for report-only motion review.
clone-websiteLoad when the user explicitly asks to clone, reverse-engineer, or rebuild a website they own or are authorized to reproduce; do not load for generic redesign, inspiration, scraping, phishing, impersonation, or copying third-party brands.
codebase-to-courseLoad when a user wants a source-backed interactive HTML course that explains how a codebase works; do not load for chat tutoring, repository review, articles, slide decks, or general browser artifacts.
context-guidanceFixture skill used to represent an additional standard skill context surface.
decision-uiLoad when Codex 即将要求用户在 2-3 个互斥方案中做真实决定、准备询问阻塞性澄清问题或需要确认高影响取舍;将选择转换为宿主原生可点击 UI,并在工具不可用时诚实文本回退。不要用于低风险实现细节、只有一个合理答案的问题、普通信息采集或不需要用户决策的完成汇报。
design-taste-frontendLoad when a task needs anti-template frontend visual direction, design-read calibration, or pre-flight critique for landing pages, portfolios, marketing pages, or redesigns; do not load for dashboards, data tables, multi-step product UI, routine frontend logic, HTML reports, slide decks, or generic code explanation.
effective-interactLoad when agent pauses to report relatively complex information needing Chinese-first clear complex communication, alignment, multi-option choice, status/incident, long-task fact ledgers, implementation plans, reviews, maps, explainers, evidence, risks, validation, handoff; choose plain text/Markdown/visual Markdown/HTML by decision cost; do not load merely because answer is long; skip trivial chat/bundled apps.