ClawHub
Back to skill

Security audit

Cursor

Security checks across malware telemetry and agentic risk

Overview

The skill is read-only, but it broadly routes Cursor-related requests through a connected OOMOL account that can expose team usage, spend, audit logs, and member data.

Install only if you intend agents to access Cursor team data through OOMOL. Use it for explicit usage, spend, audit-log, or member-list requests, and avoid enabling it in contexts where casual questions about Cursor should not touch a connected account.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger text instructs the agent to use this skill for ANY Cursor request and whenever a task involves Cursor, which is overly broad and can cause the skill to intercept loosely related prompts. In an agentic system, this can lead to unintended tool invocation, unnecessary exposure of team/account data, and bypass of more precise routing or user-intent checks.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.