skills/freedomintelligence/openclaw-medical-skills/tooluniverse-pharmacovigilance/Gen Agent Trust Hub
tooluniverse-pharmacovigilance
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches pharmacovigilance data from well-known and trusted medical services including the FDA's FAERS and DailyMed databases, PubMed, PharmGKB, and ClinicalTrials.gov. These sources are reputable for pharmacological research and the skill uses platform-specific tools to access them.
- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it processes untrusted literature from PubMed and preprint servers. Evidence Chain: 1. Ingestion points: scientific articles and trial safety signals from multiple external databases. 2. Boundary markers: absent. 3. Capability inventory: limited to generating markdown and CSV files for report purposes. 4. Sanitization: absent. The risk is considered negligible due to the restricted capability tier.
- [SAFE]: No malicious patterns such as hardcoded credentials, privilege escalation, or persistence mechanisms were detected. The skill follows standard research workflows, resolves drug identifiers through legitimate databases, and correctly implements platform-specific tool parameters.
Audit Metadata