Back to skills
extension
Category: OtherNo API key required

Skill Scout

Before you install any skill from ClawHub, run this first. It tells you if the skill is clean, flagged, or malicious — with download counts, installs, and ow...

personAuthor: gate76hubclawhub

Skill Scout

Checks a ClawHub skill before you install it. Gives you the full picture in seconds.

Why you need this

Some skills on ClawHub are flagged as suspicious or hidden by moderators. OpenClaw downloads skills without checking moderation status first — so you could install a flagged skill and not know until it's too late.

Skill Scout tells you what you're about to install before you install it.

What it checks

For any ClawHub skill, it shows:

  • Moderation verdict (clean, suspicious, malicious)
  • Reason codes if flagged
  • Download count
  • Install count
  • Star count
  • Owner username and account age
  • When it was last updated
  • License type

Trigger phrases

  • "check this skill"
  • "is [skill name] safe to install"
  • "check skill on clawhub"
  • "scan this clawhub skill"
  • "clawhub safety check"
  • "skill scout [skill slug or URL]"

How to use

Give it a ClawHub skill slug or URL. Examples:

  • "check this skill: clawhub.ai/cromabelt-watchdog"
  • "is @someuser/some-skill safe to install"
  • "check skill: someuser/cool-tool"

Output format

SKILL SCOUT REPORT
==================
Slug: someuser/some-skill
Status: ✅ CLEAN / ⚠️ FLAGGED / 🚫 MALICIOUS
Reason: [what triggered the flag]

Stats:
  Downloads: 1,234
  Installs: 567
  Stars: 42

Owner: @someuser
License: MIT-0

Recommendation: [safe to install / check the flag reason first / do not install]

Why flagged skills are a problem

  • A flagged skill may still appear on ClawHub but with reduced visibility
  • Install count gets suppressed — fewer people find it
  • Users can install it before seeing the flag
  • Malicious skills can steal data or corrupt your agent

Skill Scout runs the check before any install happens.

Requirements

  • Python 3.x
  • requests library (for API calls)