bricks
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions define a workflow where the agent is encouraged to fetch external reference URLs to inventory assets like images, videos, and SVGs to be used in the generated designs.
- Ingestion points: The agent is instructed in
SKILL.md(Workflow Step 0) andreferences/external-assets.mdto fetch user-provided reference URLs and inventory their contents. - Boundary markers: The skill does not explicitly instruct the agent to use boundary markers or provide warnings to ignore embedded instructions found on those external pages.
- Capability inventory: The skill allows the agent to generate complex JSON structures, write PHP code (hooks and custom elements), and reference arbitrary remote network assets by URL.
- Sanitization: While the skill mentions sanitization of SVG code in
references/hooks.mdand function whitelisting for PHP echo tags inreferences/dynamic-data.md, these measures are for the WordPress environment itself, not for the agent's ingestion process.
Audit Metadata