Sentinel
Sentinel is a separate Meta agent that checks Muse's outbound internet actions before they happen.
Example
Muse is about to submit a form on a website. Sentinel checks that outbound action first, and a higher-risk action comes back to you for approvalAction approval is a person's permission for one proposed action to go ahead. A useful approval shows what will happen and what it will affect..
Why it matters
Meta describes Sentinel as the layer that approves Muse's actions on the internet. Meta's safety post says read-only, previously allowed, or demonstrably low-risk actions can proceed without interruption; higher-risk actions can require your approvalAction approval is a person's permission for one proposed action to go ahead. A useful approval shows what will happen and what it will affect.. It is one part of Muse's safety design, alongside per-app permissions and an audit trail.
Common confusion
Sentinel lowers risk but does not rule it out. Meta says agent mistakes and attacks remain possible.
Sources
Updated October 3, 2026.