agentuse-share - SKILL.md Agent Skill

name: agentuse-share description: Use this skill whenever you need to add a new skill to awesome-skills, document an existing skill, create installation guides, or set up bilingual documentation. Trigger when user mentions "add skill", "document skill", "create install guide", "write skill introduction", "add to awesome-skills", or when working with the awesome-skills repository structure. Also trigger when asked to create bilingual (Chinese/English) documentation for any skill.

AgentUse Share Skill

This skill defines the complete standardized workflow for adding a new Skill to the awesome-skills repository.

Repository Directory Structure

See references/directory-structure.md for the complete directory structure and naming conventions.

Workflow

Step 1: Research the Target Skill

First, check whether a SKILL.md file is already provided within this repository at awesome-skills/skills/<skill-name>/SKILL.md.

Case A: SKILL.md Already Exists in the Repository

If awesome-skills/skills/<skill-name>/SKILL.md exists, use it as the primary source of information:

Read the SKILL.md file directly — it typically contains a comprehensive description of the skill, including its purpose, features, usage instructions, and workflow.
Supplement with the skill's GitHub repository or official documentation only if the SKILL.md does not contain sufficient information (e.g., missing supported platforms list, or lacking detail on core philosophy).
Focus on extracting the following from the SKILL.md:
- What it is: The core definition and positioning of the Skill.
- Core philosophy: What are the design principles.
- Key features and workflow: What capabilities it provides and how it operates.
- Skills library composition: What sub-skills are included.
- Supported platforms: Determine exactly which AI coding agents/platforms the Skill supports. This is critical — only create installation guides for platforms that are explicitly supported. Skip any unsupported platforms entirely.

Tip: The SKILL.md is typically more detailed and authoritative than a README. Use it as the foundation for writing the introduction documents in Steps 2 and 3 and the installation guides (INSTALL.md) in Step 4. Please refer to the content-research-writer skill's writing style to ensure the generated documentation is professional, structured, and includes clear action items. The content-research-writer skill is located at awesome-skills/skills/content-research-writer/SKILL.md.

Case B: No SKILL.md in the Repository

If no SKILL.md exists under awesome-skills/skills/<skill-name>/:

Visit the target Skill's GitHub repository or official documentation.
Use the webfetch tool (or equivalent URL fetching tool available in your environment) to fetch the README page content.
Focus on gathering the following information:
- What it is: The core definition and positioning of the Skill.
- Core philosophy: What are the design principles.
- Key features and workflow: What capabilities it provides and how it operates.
- Skills library composition: What sub-skills are included.
- Supported platforms: Determine exactly which AI coding agents/platforms the Skill supports (e.g., Claude Code, Cursor, Codex, OpenCode, OpenClaw, OpenAgent, Qoder). This is critical — only create installation guides for platforms that are explicitly supported. Skip any unsupported platforms entirely.

Step 2: Write the Chinese Introduction Document

If a SKILL.md was found in Step 1 (Case A), base the introduction on the information from that file. Summarize and restructure the content to fit the introduction template. Do NOT copy the SKILL.md verbatim — instead, distill it into a concise, reader-friendly introduction.

File path: awesome-skills/introductions/zh/<skill-name>.md
Template: See references/introduction-template.md

Step 3: Write the English Introduction Document

Same as Step 2 — if a SKILL.md exists, use it as the primary information source for the English version as well.

File path: awesome-skills/introductions/en/<skill-name>.md
Template: See references/introduction-template.md
Content should be the English translation of the Chinese version from Step 2.

Step 4: Create Platform-Specific Installation Guides

Based on the supported platforms identified in Step 1, create bilingual installation guides only for those platforms. Skip any platform that the Skill does not support.

Installation Guide Templates

CRITICAL: All installation guides MUST use bash commands that AI agents can execute. Do NOT use plugin commands like /plugin install or /plugin marketplace add because AI agents cannot execute these commands.

External Repository Skills: See references/install-guide-external-template.md
Internal Repository Skills (agent-use-skills): See references/install-guide-internal-template.md

Each installation guide should include:

Prerequisites: Required software and dependencies
Installation steps: Bash commands that AI agents can execute
Verify installation: How to confirm the installation was successful
Update method: How to update to the latest version
Getting Help: Links to GitHub repository and issue tracker

All available platforms and their corresponding file paths are documented in references/directory-structure.md.

Important: Only create rows for platforms that the Skill actually supports, as determined during Step 1 research. Do NOT create installation guides for unsupported platforms.

Step 5: Cross-Check

After all files are complete, perform the following checks:

Link validity: Ensure all installation guide links in the introduction documents have correct paths.
Bilingual consistency: Both language versions should have matching structure and content.
Information accuracy: Installation commands and steps should match the official documentation.

Important Notes

All documents must be provided in both Chinese (-zh.md) and English (-en.md) versions.
The "Installation & Support" section in introduction documents should use a hyperlinked list pointing to each platform's installation guide, rather than embedding installation commands directly.
File naming should use kebab-case (lowercase with hyphens), e.g., superpowers, ui-ux-pro-max-skill.
If the target Skill's GitHub repository cannot be accessed via browser, use the read_url_content tool to fetch page content.