ClawHub

ClawOS X Agent

Security checks across malware telemetry and agentic risk

Overview

This skill is transparently for operating the @ClawOS X account, but it enables live public posting and engagement without clear approval or credential-scope safeguards.

Install only if you intend to let an agent operate the live @ClawOS X account. Verify the `xurl` binary and authenticated account before use, constrain credentials to the intended profile, and require manual approval for every post, reply, quote, like, or retweet unless you explicitly want autonomous public engagement.

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
95% confidence
Finding
The skill description is broad enough to be invoked for many generic X/Twitter tasks, yet it controls a live brand account with posting, replying, liking, quoting, and retweeting capabilities. That increases the chance of accidental activation in the wrong context, causing unauthorized public actions, brand damage, or amplification of unsafe content.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill operates a real X account but does not prominently warn that actions can publish content and engage publicly on a live social-media profile. Without an explicit warning, users or upstream agents may treat it like a drafting-only helper and unintentionally trigger irreversible public posts, replies, likes, or retweets.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal