ClawHub
Back to skill

Security audit

Boost X Engagement

Security checks across malware telemetry and agentic risk

Overview

This skill is coherent, but it can automatically post and repost from a logged-in X account, so users should review it carefully before installing.

Install only if you intentionally want an agent to use your logged-in X session to publish public posts and reposts. Start in interactive mode, avoid scheduled mode until the config is reviewed, use low daily caps and a small repost allowlist, and do not let it write to global CLAUDE.md unless you want the behavior available in future projects.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill offers to write persistent configuration into CLAUDE.md, which modifies a local file that may affect future agent behavior across runs. Even though it asks 'Want me to write it there now?', the skill does not prominently warn that this is a persistent file modification with ongoing behavioral consequences, creating a consent and persistence risk.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.