What problem does it solve? Teams need a consistent, auditable way to decide whether a proposed action is allowed, denied, or requires approval before it executes. This Skill evaluates centralized authorization, risk, compliance, escalation, and approval policies against a proposed action and produces a deterministic decision with a full rule trace, without ever executing the governed action itself. ## Core Features & Use Cases - Deterministic Policy Decisions: Parses request context into decision attributes and evaluates authorization rules, risk and compliance constraints, then escalation and approval requirements in a fixed order. - Traceable Outputs: Emits a decision report in JSON and Markdown plus an append-only decision log, including ordered rule traces, rationale, required controls, and provenance. - Fail-Closed Behavior: Blocks or denies when policy context or decision attributes are incomplete, never producing an implicit allow. - Use Case: Before a production deployment or destructive migration, submit the action request with actor, target, scope, and mutation level to receive a deny or require-approval decision with the exact controls and approvals needed. ## Quick Start Ask the agent to evaluate whether a proposed production deployment by a specific actor is allowed, denied, or requires security approval under the repository's policy set.
