haojing8312/workclaw/skill-creator-guide

Claude Code and agent skill by haojing8312/workclaw

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Use when designing or iterating WorkClaw local skill templates and dialog-based creation flow for end users.

Install

npx vskill@latest install haojing8312/workclaw/skill-creator-guide

SKILL.md

Creating Local Skills

Overview

This built-in guide defines how WorkClaw generates local skills in a stable, maintainable way. Use it as the single source of truth for skill scaffold structure and quality checks.

Core Principles

  • Keep SKILL.md concise; include only task-relevant instructions.
  • Description must be specific and discovery-friendly, including "Use when ..." trigger conditions.
  • Use progressive disclosure: keep overview in SKILL.md and link extra files only when needed.
  • Prefer one recommended workflow instead of many equivalent options.
  • Ensure output can be verified with explicit quality checks.
  • Treat trigger quality as a first-class concern: generated skills should reduce both false positives and false negatives.
  • Keep default frontmatter minimal, but allow optional advanced fields such as allowed_tools, context, agent, and mcp-servers when the skill needs tighter runtime constraints.

Template Source

Creation Quality Checklist

  • Frontmatter includes only name and description.
  • Description is in third person and contains concrete trigger context.
  • Body includes Overview, When to Use, When Not to Use, Workflow, Prompt Examples, and Quality Checklist.
  • Generated paths use forward slashes in examples and references.
  • Default scaffold stays under a lightweight token budget.
  • Prompt examples cover both should-trigger and should-not-trigger cases.
  • Guide and template leave room for optional advanced frontmatter when needed.

Iteration Loop

  1. Observe real user-created skills and identify weak sections.
  2. Update template with minimal targeted changes.
  3. Validate generated output readability, trigger quality, and prompt examples.
  4. Look for 误触发 and 漏触发, then tighten the description or examples instead of adding verbose instructions.
  5. Keep this guide aligned with the latest template behavior.

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
155
Repository forks
1
7-day trend score
25

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/haojing8312/workclaw/skill-creator-guide/badge)](https://verified-skill.com/skills/haojing8312/workclaw/skill-creator-guide)

Something wrong with this skill or its page? Tell us.