patent-reviewer
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from patent applications and external search results. 1. Ingestion points: User-provided patent drafts and BigQuery search results processed in workflows defined in SKILL.md. 2. Boundary markers: Absent. There are no explicit instructions to ignore instructions embedded within the patent documents. 3. Capability inventory: The skill has the ability to render diagrams (render_diagram) and generate USPTO filing packages as described in SKILL.md. 4. Sanitization: Absent. The skill does not specify any filtering or validation for the processed text.
- [DYNAMIC_EXECUTION]: The skill generates diagrams by executing DOT code. Evidence: The render_diagram tool in SKILL.md accepts DOT code to create visualizations.
Audit Metadata