The Agent Skills Directory

[SAFE]: The skill instructions define a comprehensive security analyst role grounded in a specific architectural context. It establishes clear trust boundaries and identifies high-value assets to focus the AI's analysis. The core instructions include preventative security rules, such as requiring explicit user confirmation for agent actions and mandates for sanitizing inputs. The output schema is structured for technical reporting, and the workflow instructions for saving output to a session-specific directory are standard for this type of automated pipeline. No evidence of prompt injection, data exfiltration, or obfuscation was found.

phoenix-security-engineer