clinicaltrials-database
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs legitimate operations against a well-known public API and uses the standard 'requests' library. No malicious code, obfuscation, or unauthorized data access were found.
- [PROMPT_INJECTION]: An indirect prompt injection surface exists as the skill processes external text data from clinical trial records. This is assessed as safe due to the authoritative nature of the source and the lack of high-risk capabilities within the skill.
- Ingestion points: Clinical trial data retrieved from the ClinicalTrials.gov API in 'scripts/query_clinicaltrials.py'.
- Boundary markers: Absent from current script logic.
- Capability inventory: Limited to network GET requests and terminal output. No file system modification or system command execution is present.
- Sanitization: Data is used directly from the JSON responses provided by the trusted API provider.
Audit Metadata