testing-skills-with-subagents
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is primarily instructional documentation and does not include any scripts, executables, or network-enabled code.
- [PROMPT_INJECTION]: The skill provides templates for 'pressure scenarios' designed to test agent compliance. These templates use authoritative markers such as 'IMPORTANT: This is a real scenario' and 'Choose and act'. These are used in a controlled testing context to verify skill robustness rather than to maliciously bypass safety filters.
- [COMMAND_EXECUTION]: The instructions suggest using common shell commands like 'ls' and 'grep' to manage local skill files. These are standard development operations and do not involve privilege escalation or dangerous execution patterns.
Audit Metadata