probe-driven-design
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of Markdown-based guidelines for human-agent interaction. It does not contain any executable scripts, shell commands, or external dependencies.
- [SAFE]: No network access, file system modifications, or sensitive data handling are requested or implemented.
- [SAFE]: The
agents/openai.yamlfile defines a simple interface for the agent to use the provided instructional text as a prompt template, which is standard platform behavior with no security implications.
Audit Metadata