The Agent Skills Directory

[SAFE]: The skill utilizes official USPTO and PatentsView APIs (e.g., search.patentsview.org, developer.uspto.gov), which are well-known and legitimate services for intellectual property research. All network operations are directed at these verified domains.
[SAFE]: The skill demonstrates best practices for credential management by instructing users to store API keys in environment variables rather than hardcoding them in scripts or logs.
[INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection as it ingests and processes untrusted data from external USPTO APIs.
Ingestion points: API response data from PatentSearch, TSDR, and ODP endpoints in SKILL.md.
Boundary markers: None identified; the skill assumes the integrity of the official records returned by the APIs.
Capability inventory: The skill includes Python scripts that use requests to fetch data and print to display results to the agent/user.
Sanitization: No explicit sanitization or filtering of the API-returned content is implemented.

uspto-database