comparison-card
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external web searches and interpolates it into an HTML template used for automated image rendering.\n
- Ingestion points: Product parameters and comparison data are retrieved via
WebSearchas described in Step 2 ofSKILL.md.\n - Boundary markers: There are no specific instructions or delimiters to isolate the fetched data or prevent it from influencing the rendering engine.\n
- Capability inventory: The skill possesses the ability to write to the file system and execute shell commands to render graphics.\n
- Sanitization: The skill description does not include steps to sanitize or escape the data from the fetched search results before it is placed into the HTML template.\n- [COMMAND_EXECUTION]: The skill invokes a local rendering script and manages browser dependencies through shell commands.\n
- Evidence:
SKILL.mdStep 6 executespython skills/shared/scripts/render_card.py. It also notes the installation of the Playwright library and Chromium browser for the rendering environment.
Audit Metadata