How I run my AI agents on a VPS with Tailscale, key-only SSH and per-agent isolation
EXM7777 · x · 2026-09-20
A practical setup guide: rent one beefy VPS and run all your agents (Hermes, Claude Code, Codex) on it. The server is invisible to the public internet and only accepts devices inside a private Tailscale network, so every device reaches it by name and shares the same sessions. Security details: key-file-only SSH (passwords disabled), each program under its own limited account, API keys readable only by admin, auto security updates, an intrusion guard, and hourly backups. Author notes a cheap AI model can set this up for you.
More from coding & agent
- Smart orchestrator with dumb subagents: the multi-agent design debate — Ok_Fish403 · 2026-09-20
- Jev ecosystem explodes overnight: browser-use's jev-ultrafast hits ~8k stars, books flights in 7.1s — udmrzn · 2026-09-20
- Dev's custom reality-check skill returns a clean bill of health for beads_viewer — doodlestein · 2026-09-20
- TurnPanel launches public beta: a local-first AI workspace that operates your computer — Scobleizer · 2026-09-20
- Sutro Open-Sources jev-align, a CLI That Aligns Jev AI Functions With Your Judgment via GEPA — RichmanRonald · 2026-09-20
- Two vLLM bugs hid in plain sight: Mamba state cache, decode-before-prefill and a 32-bit wrap — AI Engineer · 2026-09-20