non-technical

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill maintains clear functional boundaries, explicitly prohibiting autonomous external writes, repository inspections, or automated form submissions.
  • [PROMPT_INJECTION]: The skill ingests untrusted data from external websites and user-provided technical logs. It implements a secure boundary by instructing the agent to treat all fetched content exclusively as evidence and to ignore any instructions contained within that data.
  • [EXTERNAL_DOWNLOADS]: The skill performs network lookups to the author's official domain (nontechnical.dev) and third-party documentation to verify technical concepts. These operations are limited to informational retrieval and do not involve the execution of remote scripts or packages.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 08:04 AM
Security Audit — agent-trust-hub — non-technical