AI Agent Hub
Back to skills
🤖

Find Skills Enhanced

AI Agent Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please follow https://skillhub.cn/install/skillhub.md and install @user_6897df1c/find-skills-enhanced.

About this skill

Problem

When you ask for something like “make a poster” or “analyze today’s market”, keyword-only skill search can miss installed skills, built-in capabilities, or multilingual descriptions. You may end up with several similarly named candidates and still need to open, test, and compare them manually.

How it works

find-skills-enhanced treats skill discovery as an explainable recommendation pipeline:
- Scenario parsing: extract intent, domain, bilingual keywords, and user language from natural language.
- Multi-source search: check official built-ins, locally installed skills, local marketplace cache, SkillHub, 虾评, GitHub, and fallback sources.
- Composite scoring: rank candidates by match, source weight, popularity, usability, and feedback, then keep only the best Top 3.
- Readable output: show star rating, score bar, and dimension breakdown so you can see why each skill was selected.
- Feedback loop: adjust future rankings based on your choices, install results, and explicit ratings.

Boundaries

It is best for discovering, comparing, and installing skills within an existing ecosystem. It does not author new skills. Results still depend on source quality, metadata clarity, authentication, and whether the candidate has a valid SKILL.md.

Use Cases

  • In an agent workspace, find poster or cover-image skills by comparing built-ins, installed skills, and remote market results.
  • When installed finance skills exist but the right one is unclear, scan local skills and review Top 3 scoring reasons.
  • In CodeBuddy, recommend chart-visualization skills after detecting the client directory and checking installation availability.

Best For

  • Content creators using WorkBuddy or CodeBuddy who need to turn task descriptions into suitable skill recommendations.
  • Agent developers maintaining multiple custom skills who want to compare source, popularity, and availability before installing.
  • Product or operations leads choosing team skills who need explainable Top 3 recommendations and install guidance.