principle-attack-the-premise
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides purely educational or philosophical content regarding software engineering principles and problem-solving strategies.
- [SAFE]: No network operations, unauthorized file access, or shell command executions are defined or suggested in the instructions.
- [SAFE]: The metadata and instructions do not contain any prompt injection patterns, obfuscated text, or persistence mechanisms.
- [SAFE]: The skill includes 'disable-model-invocation: true' in its frontmatter, which restricts the model's ability to execute external tools while this context is active.
Audit Metadata