ClawHub

Apify

Security checks across malware telemetry and agentic risk

Overview

This Apify skill is mostly coherent, but it advertises itself as read-focused while also enabling Apify actor execution that can create external side effects or costs.

Install only if you are comfortable letting the agent use your connected Apify account through OOMOL, including starting actor runs. Before any run_actor use, require the agent to show the actor ID, input payload, expected effect, and possible cost or external network behavior, and approve it explicitly.

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

Medium
Confidence
91% confidence
Finding
The manifest advertises the skill as limited to searching and reading data, but the documented actions include `run_actor`, which can trigger arbitrary Apify actor execution and real side effects. This mismatch can cause an upstream agent or user to trust the skill as read-only and invoke it in contexts where write/execute capabilities should have been disallowed.

Intent-Code Divergence

Medium
Confidence
90% confidence
Finding
The file contains contradictory guidance: it says the skill is for searching and reading data, but also documents `run_actor` and warns that some actions change state. In an agent setting, this inconsistency is dangerous because policy decisions may rely on the top-level description while the skill quietly exposes execution behavior deeper in the document.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger phrase 'Use this skill for ANY Apify request' is overly broad and can cause an agent to route all Apify-related tasks through a capability set that includes actor execution. That increases the chance of over-invocation, accidental use in unsafe contexts, and bypass of more restrictive workflows intended for read-only access.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal