principle-never-block-on-the-human
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains purely instructional markdown content designed to guide agent behavior. It does not include executable code, scripts, or external configuration files.
- [SAFE]: No network operations, file system access, or credential management were detected within the instructions.
- [SAFE]: The autonomy guidelines provided are balanced with clear security boundaries, explicitly stating that irreversible actions like deleting production data, force-pushing, or sending external messages still require human confirmation.
- [SAFE]: The skill uses the platform configuration field 'disable-model-invocation: true' to restrict tool usage, further limiting its potential impact surface.
Audit Metadata