principle-prove-it-works
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of markdown-formatted principles and guidelines for verifying task outputs through direct observation and repeatable scripting. It does not contain any executable code, scripts, or external dependencies.
- [NO_CODE]: No source code (Python, JavaScript, shell scripts, etc.) is included with this skill. It functions as a set of instructions for the agent to follow during its verification phase.
- [METADATA_POISONING]: The metadata fields accurately reflect the purpose and content of the skill. The use of
disable-model-invocation: trueis a restrictive configuration that limits the agent's scope during verification, which is a security-positive setting.
Audit Metadata