ClawHub
Back to skill

Security audit

ContentStudio

Security checks across malware telemetry and agentic risk

Overview

This is a broad but disclosed ContentStudio automation skill that can publish, delete, approve, moderate, and edit account data when given a ContentStudio API key.

Install only for agents you trust to act on your ContentStudio account. Use an API key with the minimum needed permissions, confirm the target workspace, and require dry-runs plus explicit approval before posting, deleting, approving, moderating, updating contacts, changing team permissions, or deleting workspace resources.

SkillSpector

By NVIDIA
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 (3)

Description-Behavior Mismatch

Low
Confidence
77% confidence
Finding
The README documents inbox moderation and contact-update actions such as hide/delete comments, delete messages, reply deletion, tag changes, and contact editing, which go beyond the narrow manifest wording of 'read and reply to DMs, comments and reviews.' In an agent context, understated scope is dangerous because users may authorize the skill expecting conversational assistance, while the tool can also perform destructive moderation or modify customer/contact records.

Vague Triggers

Medium
Confidence
91% confidence
Finding
Saying the agent 'automatically knows when to use' the CLI without defining boundaries encourages autonomous invocation of a tool that can post, approve, delete, moderate inbox items, and edit contact data. In an agent setting, ambiguous auto-use language increases the chance of the model taking externally impactful actions without explicit user intent or confirmation.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The README includes automation workflows for bulk deletion and auto-approval of posts, both of which can cause irreversible or externally visible actions at scale, but lacks prominent warnings and approval gates. In a skill meant for AI agents, copy-pastable bulk operations materially raise misuse risk because an agent may adapt and run them with little scrutiny.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.