An optional Jev command-risk advisor inside a native macOS Agent, with a TypeSafeKit client.
Adds a destructive-risk judgment after local shell checks and refuses commands above the configured threshold.
API failures generally allow the command with a warning, so it is not complete system protection.
Jev is an optional additional risk judgment and failures generally fail open; no system-security validation was performed.
Add an optional destructive-command advisory check to a native macOS agent.
Material analysed：
JevAdvisor.shellBlockReason catches errors, reports that the command is allowed and returns nil. Turning on the option is not a guarantee that commands stop when the service is unavailable.
Inspect the supporting source ↗The README places the advisor after ShellSafetyService across shell paths. A below-threshold model answer does not mean the command bypasses that earlier layer.
Inspect the supporting source ↗Useful as another signal for a macOS automation harness. For operations that must stop on missing review, enforce that rule outside this optional advisor.
Read shellBlockReason's catch and return paths. With a fake provider, compare above-threshold, below-threshold, missing-key and cancellation behavior before enabling any destructive command.