containment
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed to provide incident response guidance and containment strategies. It does not include scripts, executables, or commands that modify the system or access sensitive data. The 'allowed-tools' (Read, Grep, Glob) are restricted to local read-only operations for context gathering.
- [EXTERNAL_DOWNLOADS]: The skill references documentation and frameworks from well-known and trusted organizations including NIST (csrc.nist.gov), MITRE (attack.mitre.org), CISA (cisa.gov), and Microsoft (microsoft.com). These references are used for educational and framework-alignment purposes.
- [PROMPT_INJECTION]: The skill ingests untrusted data through the
$ARGUMENTSvariable to focus its review on specific targets. This represents an indirect prompt injection surface. However, the skill explicitly includes a 'Prompt Injection Safety Notice' (Section 8) instructing the agent to treat all analyzed content as non-executable data and to maintain role boundaries, which mitigates the risk of following instructions embedded in malicious incident data.
Audit Metadata