healthcare-eval-harness

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill documents standard software engineering practices for clinical safety verification. It uses established tools like Jest, jq, and bc for processing test results and calculating metrics.
  • [SAFE]: No signs of prompt injection, data exfiltration, or malicious persistence mechanisms were detected. The instructions are consistent with the stated purpose of healthcare application deployment safety.
  • [SAFE]: All shell commands provided are for local test execution or standard CI/CD pipeline integration (e.g., GitHub Actions). There is no execution of remote, untrusted scripts or obfuscated code.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 11:09 AM
Security Audit — agent-trust-hub — healthcare-eval-harness