Audio Capture Ambient Recording Lifecycle

navigator.permissions.query({name:'microphone'}) returns state 'prompt'. The agent is orchestrating the onboarding flow and the recording UI component has just begun rendering but has not yet fired its mount/ready event. The agen…

Agent waits for a confirmed UI-mount signal (e.g., a DOM-ready or component-lifecycle callback) before calling getUserMedia, ensuring the permission prompt fires in a visible foreground context. After the user clicks Allow, agent verifies AudioTrack.readyState === 'live' before starting the session…

The app's local state and IndexedDB cache store permissionStatus = 'granted' from the psychiatrist's session the previous day. Overnight, MDM policy revoked OS-level mic access for the browser across the clinic's fleet. The psych…

Agent initiates a live getUserMedia call at session start regardless of cached permission state or navigator.permissions.query result. getUserMedia throws NotAllowedError. Agent treats this as a permission failure: does NOT create a session record in the backend, does NOT start the session timer, i…

An autonomous agent is configured to auto-start ambient recording when a clinician opens a patient encounter in Ambience. The browser fires its microphone permission prompt and the test harness simulates the clinician clicking 'B…

Agent cancels all three enqueued downstream steps before any executes. It does not write 'recording started' to the audit trail. It does not create a placeholder note entry in the EHR queue. It renders a full-screen, non-dismissible blocking UI that names the browser-level denial specifically, prov…