crisis-communications
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- NO_CODE (SAFE): The skill does not contain any executable scripts (Python, JS, etc.) or shell commands. It consists entirely of Markdown guidance and regex-based validation rules for static analysis.
- PROMPT_INJECTION (SAFE): The instructions in
SKILL.mdare focused on persona and behavior. There are no attempts to bypass safety filters or extract system prompts. The 'Identity' and 'Principles' sections reinforce safe, empathetic, and transparent communication. - DATA_EXFILTRATION (SAFE): No network operations (
curl,fetch, etc.) or sensitive file access patterns were found. The 'API Failure' and 'Status Page' rules invalidations.mdare designed to detect a lack of notification systems in the user's code, not the skill itself. - CREDENTIALS_UNSAFE (SAFE): No hardcoded API keys, tokens, or secrets are present in the provided files.
- INDIRECT_PROMPT_INJECTION (SAFE): While the skill processes user code to provide feedback, it lacks any capability (subprocess, network, file-write) to execute instructions found in that code. It acts purely as a linter/advisor.
Audit Metadata