Running Claude Code on a Jailbroken Kindle over SSH via Tailscale

**Claude Code** is being run on a **jailbroken Kindle** reached over **SSH**, with the device joined to the operator’s network through **Tailscale**. The write-up is a public guide under the GitHub repo **Mounstroya/kindle-ssh-guide**. On Hacker News the item had **3** points and **0** comments when captured.

The stack is constrained and explicit: jailbreak the Kindle so non-stock software can run, open remote shell access with **SSH**, and use **Tailscale** so the e-reader is reachable as a peer without exposing a public SSH port. **Claude Code** is the client workload on that host; the guide is the operational path for wiring those pieces together rather than a new product release.

For engineers, the interesting bit is the host choice. A Kindle is a low-power, always-around device with a constrained UI, so putting an agentic coding CLI on it turns “thin hardware” into a remote coding endpoint if you can get a shell and a stable mesh path. That is useful for people who already live in **SSH** workflows and want a spare node that is not a laptop or cloud VM.

Market-wise this sits next to other “run AI tools on odd hardware” demos and to the broader **Claude Code** / coding-agent scene that shows up under the same HN Claude/Codex/Fable tracking. It is not a vendor announcement and, with **3** points and no thread yet, it is still a niche guide rather than a crowded product comparison.

Practical takeaway: treat the repo as the source of truth for steps and constraints; verify jailbreak, **SSH**, and **Tailscale** setup against that guide before assuming any Kindle will work. Watch whether the HN thread picks up (security of jailbroken devices on a mesh, battery and thermal limits, and whether **Claude Code** remains usable on that class of hardware) and whether similar guides appear for other e-ink or locked-down gadgets.