ClawHub
Back to skill

Security audit

Kol Account Analysis

Security checks across malware telemetry and agentic risk

Overview

This is a coherent KOL account-analysis skill that may collect public posts and comments, with disclosed public-only and de-identification limits.

Use this skill only for a clearly identified public creator/account and platform. Confirm that collection is allowed by the platform and your use case, avoid login-gated or private data, keep comment exports de-identified and retained only as long as needed, and vet any optional third-party collection tools before installing them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The quick-start prompts are broad enough that an agent could activate this skill on loosely related requests without explicit confirmation of scope, target platform, or authorization context. In a skill that drives account and comment collection/analysis, ambiguous triggers increase the chance of unintended processing of third-party data or use outside the operator's intended compliance workflow.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The README promotes collection and analysis of public account data and comment text, but the privacy/compliance warning is buried and framed as an implementation note rather than a prominent user-facing restriction. Because the skill analyzes user-generated content at scale, insufficient warning can lead operators to collect, retain, or repurpose personal data without adequate minimization, notice, or policy checks.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill explicitly instructs users to batch collect, export, and archive all public comments, videos, cover images, and post copy for influencer accounts, but it does not include any concrete privacy, retention, minimization, or lawful-basis guidance. Even though the data is described as public, large-scale aggregation and export of user comments can materially increase privacy risk, create compliance exposure, and normalize excessive collection without clear safeguards.

Natural-Language Policy Violations

Medium
Confidence
89% confidence
Finding
The file explicitly requires Chinese-language output in Step 1 deliverables without any user language choice or opt-in. This can override user preferences or higher-level application expectations, creating prompt-quality and accessibility issues, though it is not a direct code-execution or data-exfiltration risk.

VirusTotal

47/47 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.