uspto-database

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill retrieves patent records and metadata from the official PatentsView API (api.patentsview.org) and Google Patents Public Data on BigQuery. These are well-known and reputable services for accessing public patent information.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external data (patent titles and abstracts) that could theoretically contain malicious instructions.
  • Ingestion points: External data is ingested via requests.post calls to the PatentsView API in SKILL.md.
  • Boundary markers: The code examples do not implement specific boundary delimiters or safety instructions to the agent to ignore potentially malicious content within the retrieved patent text.
  • Capability inventory: The skill includes file writing capabilities via matplotlib (plt.savefig) and data processing through pandas and requests.
  • Sanitization: There is no explicit sanitization or filtering of the retrieved API data before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 03:35 PM
Security Audit — agent-trust-hub — uspto-database