If you have ever tried to pick up a local coding session from your phone or the browser and hit the message “remote control is disabled by your organization’s policy”, you have run into one of Claude Code’s centrally enforced settings. This is not a bug you can fix in your own terminal — it is a switch only an organization owner can flip. Here is exactly what the feature does, how to turn it on, and the two edge cases that still bite after you do.
What Claude Code Remote Control Actually Does
Remote Control is a Claude Code feature that lets you continue a local coding session from another device. You start the session on your machine, then steer it from claude.ai/code or the Claude mobile app.
The important detail for security-conscious teams: the session still runs entirely on the original machine. Your agent keeps full access to the local filesystem, tools, and project configuration. What syncs to the remote viewer is the transcript — so you can watch and guide the session, but the work never leaves your laptop. That is the model that makes this acceptable in enterprise environments where code cannot leave the perimeter.
On Team and Enterprise plans, Remote Control is off by default. Until an administrator turns it on, anyone who runs the remote-control command gets the policy error above.
How an Admin Enables It
If you are the organization owner (or have admin rights), the fix is a single toggle:
- Open claude.ai/admin-settings/claude-code.
- Find the Remote Control toggle and turn it on.
- Save.
The setting then shows as “Set by admin” — individual team members can no longer turn it off for themselves. This is intentional: it is an organizational policy, not a per-user preference.
When the Toggle Is Greyed Out
If the toggle is disabled and you cannot click it, your organization has a data-retention or compliance setting (such as Zero Data Retention) that is incompatible with Remote Control. That constraint cannot be changed from the panel — you would need to contact Anthropic support to resolve it. Do not assume the feature is broken; the greyed-out state is the compliance guardrail doing its job.
The Trusted Devices Layer (Beta)
Once Remote Control is on, you can add a second gate: Require trusted devices. When enabled:
- Members must verify each new device (Face ID, Touch ID, Windows Hello, or a passkey) before it can view or control a session.
- The device must stay signed in within the last 18 hours.
- It applies only to sessions started after you turn it on — anything already running is not retroactively enrolled.
Worth a heads-up to your team before flipping it, because the first remote interaction on each device will trigger enrollment.
The Per-Device Override That Still Bites
Here is the edge case that trips up even teams that have enabled the org toggle. Remote Control can also be blocked at the individual device level through a disableRemoteControl managed setting, which is common on MDM-managed machines.
If your organization pushes that policy, the org toggle being “on” will not matter for that specific device — the device-level rule wins. In that case a separate device policy has to be lifted by whoever manages the MDM, independent of the Claude admin panel.
So the resolution path is two-layered:
- Org level — owner enables Remote Control at
claude.ai/admin-settings/claude-code. - Device level — MDM admin removes the
disableRemoteControlmanaged setting on the affected machine.
How Team Members Connect
Once the admin has saved the setting, anyone who was hitting the error just needs to run the /remote-control command again inside a Claude Code session (or restart the session). It should now connect and hand back a session URL that works at claude.ai/code or in the mobile app.
Quick Checklist
- Seeing the policy error? → You are not the owner, or the toggle is off. Ask your org admin.
- Owner, toggle greyed out? → Compliance setting (e.g. Zero Data Retention) blocks it — contact Anthropic support.
- Toggle on but still blocked on one machine? → MDM
disableRemoteControlis overriding it — lift the device policy. - Want extra security? → Enable Require trusted devices and warn your team about first-device enrollment.
Remote Control is a genuinely useful feature for developers who move between a workstation and a phone, and the security model — local execution, synced transcript — is sound. Just remember that the error is telling you the truth: it is an organization policy, and only the right administrator at the right layer can change it.
