ClawHub

frigatebird

Security checks across malware telemetry and agentic risk

Overview

This skill is not malicious, but it lets an agent use a live X account through browser cookies and perform public account-changing actions without enough guardrails.

Install only if you are comfortable letting an external CLI act as your logged-in X account. Use a dedicated browser profile or test account, avoid putting raw cookies or tokens in prompts or logs, verify the npm package source/version, and require explicit confirmation before posting, replying, following, liking, retweeting, or changing lists.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (2)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill explicitly promotes operating on X via browser session cookies and supports account mutations, but it does not warn users that supplying auth tokens, ct0 values, or browser profiles grants the tool the ability to act as their account. In an agent setting, missing consent and safety language increases the chance of unintended posting or account misuse.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The documented commands include tweeting, replying, publishing articles, following/unfollowing, and modifying lists, all of which change live account state or relationships. Without an explicit warning, dry-run guidance, or confirmation requirement, an agent may execute destructive or reputation-impacting actions on a real account.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal