competitive-ads-extractor

Fail

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: HIGHPROMPT_INJECTIONNO_CODE
Full Analysis
  • PROMPT_INJECTION (HIGH): The skill is highly vulnerable to indirect prompt injection due to its design of ingesting untrusted external data (ad copies from Facebook and LinkedIn) and performing high-privilege operations like writing to the local filesystem. 1. Ingestion points: External ad libraries. 2. Boundary markers: Absent in instructions. 3. Capability inventory: Web scraping and local file writing (~/competitor-ads/). 4. Sanitization: None described. A malicious actor could embed commands in an ad that the agent might execute while processing.
  • NO_CODE (INFO): The skill contains only documentation and instructions in a Markdown file with no provided source code or executable scripts, making it a purely instructional skill for an agent to interpret.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 15, 2026, 10:25 PM