The Agent Skills Directory

[SAFE]: The skill defines a logical, procedural workflow for infrastructure deployment and CI/CD operations without any suspicious command execution or network exfiltration patterns.
[SAFE]: Implements a 'Socratic Gate' that requires the agent to verify target environments, test results, and breaking changes before proceeding, ensuring human-in-the-loop safety.
[SAFE]: Contains explicit rules to protect sensitive information, specifically stating 'Never expose secrets'.
[SAFE]: The use of a local memory file (.ai-memory.md) for state persistence is a standard pattern for context management and does not involve unauthorized data access or exfiltration.

deploy