clinpgx-database
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates data retrieval from the external endpoint
https://api.clinpgx.org/v1/. This connection is essential for the pharmacogenomic database functionality described in the overview. - [PROMPT_INJECTION]: The skill processes structured data from an external API, creating a surface for indirect prompt injection if the source were compromised.
- Ingestion points: Data enters the context through JSON responses from
api.clinpgx.orgvia the functions defined inscripts/query_clinpgx.py. - Boundary markers: The instructions do not specify any delimiters or safety markers to isolate external API content from the agent's internal reasoning logic.
- Capability inventory: The skill includes the ability to write to the local filesystem (for caching data via the
cached_queryfunction) and initiate network requests. - Sanitization: There is no evidence of data sanitization or validation of the API's JSON payload before the content is presented to the agent.
- [SAFE]: Analysis of the source code and documentation found no evidence of credential theft, hidden communication, or malicious intent. The skill utilizes standard, verifiable libraries and communicates with a well-known scientific data provider.
Audit Metadata