[Feature Request] Voice input (mic button / dictation) in Mobile Remote Control — and a native iOS companion app

Summary

Requesting voice input in the Mobile Remote Control web UI, and longer term, a native iOS companion app (the way Claude Code ships Remote Control inside the Claude iOS app, and Copilot CLI sessions are steerable from the GitHub mobile app).

Environment

• ZCode desktop on Windows 11 (10.0.26200), current stable
• Client: iPhone, Safari, connected via Mobile Remote Control (QR pairing)

Current behavior

1. Open ZCode desktop, start Remote Control (phone icon, lower-left sidebar).
2. Scan the QR code from the iPhone; the workspace opens in mobile Safari.
3. The prompt input in the session view is text-only: no mic button, no dictation integration. No voice setting exists anywhere in Settings or the docs, and no marketplace plugin provides one.

Expected behavior

• A mic button in the session-view input on the mobile Remote Control page (the Web Speech API / on-device dictation works in iOS Safari, so no cloud STT dependency is required).
• Optionally, spoken (TTS) agent replies for a fully hands-free loop.
• Longer term: a native iOS companion app reusing the existing Remote Control protocol.

Why phone-side input matters

Desktop-side voice via MCP servers (e.g. Whisper-based voice servers) captures from the desktop microphone — which doesn't help when the user is away from the desk controlling from the phone. The mic has to be on the phone side of the connection.

Additional context

• Current workaround: iOS keyboard dictation into the prompt box. It works, but a native push-to-talk button would be much smoother.
• Related demand on other harnesses: anthropics/claude-code#25115 and anthropics/claude-code#29399 (voice input for Remote Control) show this is a cross-harness pain point, and neither has shipped it — a good chance to lead here.

Start with the Mobile Remote Control session view and its prompt input, then review the existing Remote Control protocol and QR-pairing flow. Check how iOS Safari handles the Web Speech API and identify the scope difference between web voice input, spoken replies, and a native iOS companion app. Done requires an agreed scope and acceptance criteria for the selected feature.