ns-agent-adaptation
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows a strict interview protocol to assist users in designing AI agents. It does not include any executable code or framework recommendations, adhering to its stated purpose of conceptual design.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text during its 'grill-me' interview turns and writes the resulting design to a local file (
docs/specs/agent-design.md). While this creates a surface for indirect prompt injection, the risk is negligible as the skill maintains a rigid interaction schema and the output is limited to a specific documentation path. - [REMOTE_CODE_EXECUTION]: No remote code execution patterns were detected. The skill specifically forbids recommending or implementing code or frameworks.
- [DATA_EXFILTRATION]: No data exfiltration patterns or network operations were found. The skill operates locally within the project's file system structure.
- [CREDENTIALS_UNSAFE]: No hardcoded credentials, API keys, or sensitive environment variable patterns were identified.
Audit Metadata