ClawHub
Back to skill

Security audit

Buffett-Style Moat Analyzer

Security checks across malware telemetry and agentic risk

Overview

This appears to be a disclosed finance-analysis skill with some wording that should be treated carefully, but no evidence of hidden execution, persistence, credential use, or data exfiltration.

Install only if you want informational business or investment-risk analysis, and treat outputs as educational rather than personalized financial advice. Do not rely on it for buy, sell, hold, price target, legal, tax, or suitability decisions without independent verification and a qualified professional where appropriate.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Description-Behavior Mismatch

Medium
Confidence
91% confidence
Finding
The manifest description says the skill 'must not provide financial advice, price targets, or buy/sell recommendations.' However, L101 says not to use terms like 'buy', 'sell', 'hold', 'undervalued', or 'overvalued' unless the user asks for a valuation model and reliable data is available, which broadens the skill beyond the manifest's stated prohibition. This is a semantic mismatch between the declared safe scope and the documented behavior.

Vague Triggers

Low
Confidence
88% confidence
Finding
This markdown file defines when to use the reference with the phrase 'when the user asks for a harsh review, downside analysis, or acquisition/stock risk screen.' Terms like 'harsh review' and 'downside analysis' are fairly broad and lack explicit boundaries or exclusion examples, which could cause the skill to be invoked for general critique requests outside the intended investment/business context.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.