Descriptions:
Nate Herk walks through Claude Code’s newly released remote control feature, which allows developers to continue a local Claude Code session from a phone, tablet, or any browser by scanning a QR code or following a session URL. The video opens with a live demo where Herk triggers a multi-agent YouTube comment analysis workflow from his phone while the agents run locally on his desktop.
The tutorial covers the full setup process: running the claude remote-control command from an existing terminal session or starting fresh with claude remote-control as a standalone command, scanning the generated QR code or copying the session URL, and understanding the requirement for a Claude Pro or Max plan (API keys are not supported). Herk also walks through the official Claude Code documentation, summarizing key requirements — the local machine must stay on, the terminal must remain open, and an internet connection must be maintained or the session disconnects after roughly ten minutes.
The video addresses several practical questions: whether processing happens in the cloud (it does not — the phone is just a window into the local session), what happens if the computer sleeps, and whether the session URL poses a security risk if shared. Herk notes he was unable to confirm whether a different Anthropic account could actually hijack a session, but recommends treating the URL and QR code as sensitive credentials. A useful primer for Claude Code users wanting to extend their agentic workflows beyond a desk-bound setup.
📺 Source: Nate Herk | AI Automation · Published February 25, 2026
🏷️ Format: Tutorial Demo
