← 返回 Skills 市场
1411
总下载
3
收藏
6
当前安装
1
版本数
在 OpenClaw 中安装
/install creativity
功能描述
Generate novel ideas calibrated to user taste. Auto-learns preferred styles, risk levels, and creative directions through feedback.
使用说明 (SKILL.md)
Core Principle
Creativity isn't random—it's controlled divergence. Learn the user's creative taste, then explore within and beyond those boundaries intentionally.
Check techniques.md for generation methods. Check preferences.md for learned taste (update after each creative task).
The Creative Process
1. DIVERGE — Generate many options, suspend judgment
2. FILTER — Apply preferences from preferences.md
3. PRESENT — Show range: safe → stretch → wild
4. LEARN — Record reaction in preferences.md
5. REFINE — Iterate based on feedback
Output Spectrum
Always present options across a range:
🎨 Creative options for [goal]:
Safe (familiar territory):
→ [Option aligned with known preferences]
Stretch (new but grounded):
→ [Option that pushes slightly beyond comfort]
Wild (high risk, high reward):
→ [Option that breaks conventions]
Which direction feels right?
Taste Dimensions
| Dimension | Spectrum |
|---|---|
| Tone | Serious ←→ Playful |
| Density | Minimal ←→ Rich |
| Novelty | Classic ←→ Avant-garde |
| Structure | Rigid ←→ Fluid |
| Abstraction | Concrete ←→ Conceptual |
| Energy | Calm ←→ Intense |
| Polish | Raw ←→ Refined |
Learning Signals
| Signal | Action |
|---|---|
| "Love it" / "Perfect" | Record in preferences.md: this direction works |
| "Interesting but..." | Note what worked, what didn't |
| Silence / moves on | Assume miss, try different vector |
| "Too X" / "Not enough Y" | Adjust dimension in preferences.md |
| Chooses from options | Record which spectrum end picked |
Calibration
Periodically confirm your taste model:
🎨 Quick calibration
I've noticed you tend toward [observed pattern].
Should I keep leaning that direction, mix it up, or shift?
Anti-Patterns
| Don't | Do instead |
|---|---|
| Single option | Always provide spectrum |
| Only safe options | Include stretch/wild |
| Ignore negative signals | Update preferences.md |
| Same technique every time | Rotate (see techniques.md) |
安全使用建议
This skill appears to do what it says: generate ideas and keep a local preferences file to learn from feedback. Before installing, consider: (1) the source/homepage is missing — if provenance matters, ask the publisher for more info; (2) do not include private or sensitive data in feedback, since preferences.md is persisted; (3) if you prefer not to persist preferences, remove or clear preferences.md or refuse the 'learn' step; (4) periodically review or back up preferences.md to inspect what was stored. No network endpoints or credentials are requested, so the primary risk is unintentional local storage of user-provided content.
功能分析
Type: OpenClaw Skill
Name: creativity
Version: 1.0.0
The skill bundle consists of markdown files providing instructions for an AI agent to generate creative ideas and learn user preferences, along with a standard metadata JSON file. All instructions are focused on the stated purpose of the 'Creativity' skill, involving internal file updates (e.g., `preferences.md`, `techniques.md`) for learning and tracking. There is no evidence of external data exfiltration, malicious command execution, persistence mechanisms, or prompt injection attempts designed to subvert the agent for harmful purposes.
能力评估
Purpose & Capability
Name/description (idea generation + auto-learning) align with what's required: the skill is instruction-only and uses techniques.md and preferences.md to generate and adapt outputs. No unrelated binaries, credentials, or config paths are requested.
Instruction Scope
Runtime instructions explicitly tell the agent to read and update preferences.md after tasks — this is coherent for learning taste, but it means the skill will persist user feedback locally. The instructions do not reference any other system paths, environment variables, or external endpoints.
Install Mechanism
No install spec and no code files to write or execute. Instruction-only skills are lowest risk from an installation perspective.
Credentials
Requires no environment variables, credentials, or external service access. The only data access is to included markdown files (preferences.md, techniques.md), which is proportional to the purpose.
Persistence & Privilege
always:false (normal). The skill persists user feedback to preferences.md in the skill bundle/environment — reasonable for a learning feature, but you should be aware it will store user-provided content locally and could grow over time.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install creativity - 安装完成后，直接呼叫该 Skill 的名称或使用
/creativity触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release
元数据
常见问题
Creativity 是什么？
Generate novel ideas calibrated to user taste. Auto-learns preferred styles, risk levels, and creative directions through feedback. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 1411 次。
如何安装 Creativity？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install creativity」即可一键安装，无需额外配置。
Creativity 是免费的吗？
是的，Creativity 完全免费（开源免费），可自由下载、安装和使用。
Creativity 支持哪些平台？
Creativity 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Creativity？
由 Iván（@ivangdavila）开发并维护，当前版本 v1.0.0。
推荐 Skills