The Agent Skills Directory

[SAFE]: The skill implements a 'Workflow Engine' that governs how AI agents interact during development. It emphasizes security by requiring a 'SECOPS_APPROVED' gate for any security-sensitive changes (auth, secrets, PII, etc.), regardless of the change size. All operations involve local file management (reading workflow.yaml or .workflow-state.json) or documented optional MCP integrations for task management. No unauthorized network access, credential harvesting, or obfuscation was detected.

workflow-engine