advance-health-care-directive

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill is entirely declarative and does not contain any executable logic or command-line operations.
  • [NO_CODE]: No Python or Node.js code is included in the skill; it functions as a prompt-based template for structured information gathering and document drafting.
  • [PROMPT_INJECTION]: The content was analyzed for override markers and bypass attempts; none were found. The instructional language is professional and focused on the stated legal drafting purpose.
  • [DATA_EXFILTRATION]: There are no network requests, API calls, or file system writes. While the workflow requests sensitive PII/PHI (names, medical history), the skill provides no mechanism to transmit this data outside of the agent's context.
  • [EXTERNAL_DOWNLOADS]: No external URLs or remote package downloads are referenced for execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 07:55 PM