support-deflection-knowledge-base-review
Pass
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed of Markdown and YAML configuration files providing domain-specific guidance for customer support operations. It does not contain any executable scripts, binary files, or network-enabled tools.
- [PROMPT_INJECTION]: No evidence of malicious prompt injection, role-play bypasses, or instructions to ignore safety guidelines. The skill includes positive safety guardrails to ensure users are not trapped in automated systems and have access to human help for high-stakes issues.
- [DATA_EXFILTRATION]: No network operations or commands to access sensitive local files were found. Rule
support-deflection-7explicitly instructs the agent to collect diagnostics 'without exposing secrets'. - [INDIRECT_PROMPT_INJECTION]: While the skill processes user-supplied support context and questions, it lacks high-risk capabilities such as shell execution, file system writes, or network access that could be exploited via malicious input. The risk is limited to the accuracy of the support deflection plan generated.
Audit Metadata