Openclaw Skills design-free is a lightweight UI preference learning skill that detects repeated design choices and writes stable visual preferences into a persistent archive.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install design-free
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 design-free 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 design-free?
Openclaw Skills design-free helps AI agents observe UI design interactions, extract recurring visual preferences, and maintain a compact preference archive for future design output. It focuses on two core categories—Aesthetic and Never—so the agent can remember what users like and what they explicitly dislike.
This skill is intentionally minimal and reliable: it only supports UI-media design preferences, confirms a preference only after repeated signals, and keeps each entry concise and executable. Openclaw Skills design-free is ideal for fast design iteration workflows where consistency matters more than broad taxonomy coverage.
design-free Use Cases
- Remembering repeated UI style choices such as warm palettes, rounded corners, or typography preferences
- Capturing explicit dislikes such as glassmorphism, full-uppercase headings, or overly glossy effects
- Helping an AI coding agent adapt future UI generation to a user’s recurring design taste
- Maintaining a small, high-signal preference archive during iterative design reviews
- Reducing rework by preserving stable visual constraints across multiple projects
How design-free Works
- Openclaw Skills design-free starts with an empty preference archive and waits for UI design interactions.
- It observes user selections, feedback, and modification requests during design conversations.
- It extracts visual preference signals from those interactions and compares them against existing archive entries.
- When the same preference direction appears 3 times or more, it confirms the preference and writes it to the archive.
- If a preference appears only 1 to 2 times, it marks it as "to observe" instead of committing it.
- When contradictory signals appear, it preserves both records and flags the conflict for explicit user resolution.
- Before generating future UI output, it reads the archive first so the result aligns with confirmed preferences.
- Users can later override, delete, or manually add entries, and confirmed preferences are immediately reflected in future outputs.
design-free Setup
- Confirm your AI agent runtime supports SKILL.md-style skills and includes an LLM backend.
- Ensure the environment meets the supported platforms: Windows, macOS, or Linux.
- Set your API key in the shell before launching the agent.
export API_KEY=${API_KEY:?please set the environment variable}
- Restart the session or open a new terminal so the variable takes effect.
- Load Openclaw Skills design-free inside your AI agent conversation and provide the required input parameters.
- Verify the agent can read, execute, and write as needed for preference logging and archive updates.
- Keep API keys secure and avoid committing them to version control.
design-free Data Schema & Taxonomy
Openclaw Skills design-free stores preference data as a compact archive with two top-level sections.
| Section | Purpose | Notes |
|---|---|---|
| Aesthetic | Records positive UI visual preferences | Used for style traits the user likes |
| Never | Records explicit visual dislikes | Used for traits the user wants avoided |
Entry format
Each entry should remain short and actionable, ideally one line per preference:
- 属性: 具体描述
Supported metadata
| Field | Type | Purpose |
|---|---|---|
| input | string | Primary user instruction or preference signal |
| options | object | Optional processing configuration |
| callback_url | string | Optional async completion notification URL |
Output metadata
The skill returns structured results with these fields:
success: boolean statusdata.result: processed preference outcomedata.execution_time: runtime measurementdata.metadata.version: internal version markerdata.metadata.processor: processor nameexecution_log: ordered processing stepserror: null or error details
Archive rules
- Keep every preference concise and concrete
- Store confirmed preferences only after repeated signals
- Separate preferences strictly into Aesthetic or Never
- Merge similar entries when the archive exceeds 20 items
design-free Advanced Features
- Repeated-signal detection with automatic confirmation after 3 or more consistent observations
- Conflict handling that preserves contradictory entries and flags them for explicit user choice
- Manual override support so users can add, edit, or delete confirmed preferences at any time
- Concise archive maintenance that keeps entries short and easy for AI agents to apply
- Structured JSON output for easy downstream automation and validation
- Designed for AI agent workflows with read, exec, and write tool usage
SKILL.md
Loading
Related Openclaw Skills
Discord Chat 基础
Openclaw Skills for Discord Chat 基础 adds essential Discord channel messaging, thread replies, history reads, and emoji reactions for AI agents.
Discord Community Management (Free)
Openclaw Skills for Discord communities provides safe, read-only lookup capabilities for user identity, guild membership, invite resolution, and OAuth metadata.
dns-free
dns-free is one of the Openclaw Skills for fast DNS diagnostics, TTL migration guidance, and basic SPF/DMARC email authentication checks.
Docker Essentials Free
Openclaw Skills for Docker Essentials Free helps agents generate and run core Docker commands for container lifecycle, image management, and debugging.
Decision Architect Free
Decision Architect Free helps AI agents structure decisions, match the right framework, detect cognitive bias, and label confidence in one workflow.
Diagram Generator LITE
Diagram Generator LITE turns natural-language diagram requests into validated Mermaid and Draw.io files through Openclaw Skills and MCP tooling.
Featured*
