evidence-synthesis-ai

Pass

Audited by Gen Agent Trust Hub on Apr 26, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill does not contain any malicious instructions, obfuscation, or unauthorized commands.- [EXTERNAL_DOWNLOADS]: The skill documents the use of external biomedical databases (PubMed/MEDLINE, Cochrane CENTRAL) which are standard for clinical research. No unverified remote code execution or suspicious script downloads were identified.- [DATA_EXFILTRATION]: No evidence of sensitive local file access or exfiltration to unauthorized domains. Network operations are limited to clinical database queries.- [PROMPT_INJECTION]: The skill ingests external research data which presents a minor surface for indirect prompt injection (Category 8). Evidence: Ingestion occurs via PubMed/Cochrane data; boundary markers are not explicitly defined in the text; capability inventory is limited to text generation; and no specific sanitization logic is described. However, this is inherent to the skill's purpose and does not elevate the risk level.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 26, 2026, 10:56 PM
Security Audit — agent-trust-hub — evidence-synthesis-ai