AEO Prompt Research (Free)

• Outcome: Pass
• Latest audit: May 28, 2026 at 11:39 AM UTC
• Version:
  1.0.0

Overview

This skill appears to fetch user-selected websites for prompt research, and the shell-based crawl path is disclosed and purpose-aligned, though users should be aware it makes outbound web requests.

Install only if you are comfortable with the agent making outbound requests to domains you provide. Prefer using it on public sites, avoid internal or private network hosts, and review the crawl script/metadata if you need strict control over command execution.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 84% confidence
• Finding: The skill explicitly instructs the agent to run `scripts/crawl_site.sh <domain>` as an alternative crawl path, which introduces shell execution capability that is not declared in the skill metadata or requirements. Undeclared shell use is risky because it expands the attack surface and, if the domain input is not safely handled by the script, could enable command injection or unexpected local command execution.

VirusTotal

64/64 vendors flagged this skill as clean.