ClawHub
Back to skill

Security audit

X (Twitter)

Security checks for vulnerabilities and agentic risk

Overview

This skill is a clearly disclosed X/Twitter connector that can read and change account data through OOMOL, with write and destructive actions called out for confirmation.

Install only if you are comfortable letting OOMOL-connected tooling operate your X/Twitter account, including reading private account data such as DMs when requested. Review payloads carefully before approving posts, follows, DMs, list changes, media uploads, or deletions.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Intent-Code Divergence

Medium
Confidence
91% confidence
Finding
The safety section tells the agent that untagged actions are safe reads, but the action list contains inconsistent tagging and ambiguous operations. That can cause an agent to execute state-changing, privileged, or costly actions without confirmation because it relies on incorrect in-file safety classification rather than the live schema or an allowlisted risk map.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The instruction to use this skill for ANY X request is overly broad and encourages unconditional routing to a highly privileged integration. In an agent setting, broad triggers increase the chance of invoking write/destructive capabilities in contexts where a narrower skill, safer read-only path, or additional policy checks should apply.

Static analysis

No suspicious patterns detected.