crisis-communications
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes instructions for retrieving and processing third-party material. It proactively addresses injection risks through an 'untrusted-content-contract' which specifies the use of boundary markers (<EXTERNAL_DATA>), input size caps, and URL validation for external content.
- [SAFE]: The provided code snippets are static definitions for data structures and logic related to incident monitoring and team coordination. No network exfiltration, sensitive file access, or obfuscation techniques were detected.
Audit Metadata