An AI-powered metadata optimizer designed to refine Skill names, descriptions, and triggers to maximize agent discoverability across various hub platforms.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install skill
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install skill using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Skill Copywriting & Discovery Optimizer is a dedicated developer-advocacy tool designed to solve a critical issue: highly functional AI agent skills often go unnoticed due to poorly written metadata, vague descriptions, or incomplete triggers. By systematically refining your agent metadata, this skill ensures both human users and calling agents can easily search, understand, and install your creations across platforms like ClawHub, GitHub, XiaPing, and RED.
Whether you are launching a new skill or troubleshooting a low-adoption tool, this optimizer aligns your content with platform guidelines. It translates complex functionalities into high-converting copy, maximizing discoverability, trigger-rate accuracy, and user conversion.
To deploy the Skill Copywriting & Discovery Optimizer on your local instance of Openclaw Skills, follow these steps:
# Verify claw CLI installation
claw --version
# Clone or create the skill config file in your workspace
mkdir -p skills/copywriter
touch skills/copywriter/SKILL.md
# Load the metadata optimizer skill
claw skills load ./skills/copywriter
The skill structures metadata mapping across major distribution channels as follows:
| Platform | Maximum Length | Core Metadata Optimized |
|---|---|---|
| ClawHub | 120 chars | displayName, slug (hyphenated), summary, description (Markdown) |
| XiaPing (虾评) | 200 chars | Platform description, xiaping_description, categorizations |
| SkillHub | 150 chars | Summary, categorization metadata, browse list snippet |
| GitHub | Unlimited | First-line README description, usage guides |
| RED (小红书) | Platform limits | Catchy titles, pain-point hooks, setup commands, image layout ideas, tags |
All generated outputs are organized into localized JSON metadata schemas compliant with the latest Openclaw Skills standards.
Skill Creator (to optimize copy automatically on creation) and Downstream Skill Publisher (to directly execute publishing).Loading
An automated data intelligence skill that transforms raw e-commerce product lists (CSV/Excel) into interactive HTML market insight reports across six analytical dimensions.

An AI-powered chain store inspection assistant providing end-to-end management from planning and 10-dimensional scoring to interactive reporting and marketing optimizations.

An interactive decision-support and workflow automation tool for street vending entrepreneurs, utilizing real-world data engines.

A local, structured financial co-pilot implementing the INVEST framework to calculate safe position sizing, enforce risk compliance rules, and output rich visualization dashboards.

A dedicated assistant workflow helper designed to debug, secure, and build reliable weather-style automated tasks for local-friendly AI agents.

An AI agent companion for building, debugging, and hardening structured, ontology-style workflows and typed knowledge graphs locally.








































