pricing-page-psychology-audit
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external websites for analysis, creating a vulnerability to instructions hidden within the scraped content (e.g., hidden text or HTML comments intended to derail the AI's logic).\n
- Ingestion points: The
scripts/scrape_pricing.pyscript fetches full text, plan names, and button labels from user-provided URLs.\n - Boundary markers: There are no explicit instructions in
SKILL.mdfor the agent to ignore or delimit potentially malicious instructions within the scraped data.\n - Capability inventory: The skill executes shell commands (
python) and can perform various analysis and reporting tasks based on the input.\n - Sanitization: The scraper performs basic whitespace cleanup but does not sanitize or filter the content for injection patterns.\n- [COMMAND_EXECUTION]: The skill uses a local Python script (
scripts/scrape_pricing.py) to perform its primary data collection task. This script is bundled with the skill and executes as part of the intended workflow.\n- [EXTERNAL_DOWNLOADS]: The scraping script makes outbound network requests to fetch HTML content from URLs provided by the user. While this is necessary for the skill's function, it allows for interaction with arbitrary external domains.
Audit Metadata