hard-news
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional and conversational. It does not involve any code execution, file system access, or network communication. It functions purely as a prompt template for the AI agent.
- [SAFE]: No obfuscation, prompt injection, or malicious patterns were detected in the instructions. The behavior is consistent with the described purpose.
- [SAFE]: The skill configuration specifies
disable-model-invocation: true, which prevents the agent from calling any external tools, effectively eliminating common attack vectors related to tool-based exfiltration or execution.
Audit Metadata