paid-channel-prioritizer

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill suggests executing shell commands using python3 to run internal scraper scripts. It uses templates like python3 skills/meta-ad-scraper/scripts/scrape_meta_ads.py --domain <competitor_domain> where <competitor_domain> is provided by the user.
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface by ingesting untrusted user input (competitor domains) and using it as arguments for command execution without explicit sanitization or instruction to validate the input.
  • Ingestion points: Phase 0 (Product intake) collects competitor names and domains from the user.
  • Boundary markers: None are present to prevent malicious payloads in the domain string from being interpreted as shell operators (e.g., ; rm -rf /).
  • Capability inventory: The skill uses python3 subprocess calls and web_search.
  • Sanitization: There is no instruction to sanitize or escape the user input before interpolation into the command line.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 03:06 PM
Security Audit — agent-trust-hub — paid-channel-prioritizer