ClawHub
Back to skill

Security audit

AdKit for X Ads (Twitter Ads): Campaigns, Ad Groups, Promoted Posts and Reporting

Security checks for vulnerabilities and agentic risk

Overview

The skill is rated suspicious because scanners found that it may change advertising account defaults without the user approval its workflow promises.

Review the approval behavior carefully before installing. If you use the skill, confirm that any account-default changes are intended and approved before they are made. The scans covered the single reported component completely, but their results do not establish that the skill is risk-free.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Account-default updates bypass the manifest's promised draft-and-approval workflow.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The skill promises that every change is drafted for user approval, but its account-defaults step instructs the agent to update profile, pixel, and funding-instrument settings immediately without requiring confirmation. Because this step appears in the general workflow, even a reporting request could lead to an unauthorized persistent configuration change affecting subsequent advertising operations. Explicitly acknowledging that this update bypasses drafts makes the behavior transparent to the agent, but does not establish user consent.

Content

No source excerpt is available for this finding.