ClawHub
Back to skill

Security audit

X use

Security checks across malware telemetry and agentic risk

Overview

The skill appears intended for Twitter/X collection, but it combines account-scoped reads with automatic local retention in a way users should review before installing.

Install only if you intentionally want Twitter/X content collected into a local knowledge base. Before using it, confirm which commands access your logged-in account, where collected tweets are stored, how to delete retained data, and require explicit confirmation before collecting private or personalized timeline/feed content.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger description is broad enough to activate on many ordinary Twitter/X-related requests, increasing the chance the agent invokes this skill when the user did not intend browser-backed collection or account-context actions. In this skill, unintended activation is more dangerous because commands may access a logged-in Twitter session and can fetch personalized timeline data, creating privacy and overreach risks beyond a simple search tool.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill states that collected tweets are automatically stored in a local knowledge base, but this persistence is not disclosed in the top-level description that drives selection and user expectations. That creates a transparency and privacy issue: users may believe the tool performs a one-time fetch, while it actually retains potentially sensitive personalized timeline, reply, or search data for later querying.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal