Watch the agent think.
Every command, every file change, every reasoning step — streamed live as the agent writes them. Vibe coding, with receipts.
self-hosted · for solo developers
Yours doesn't need to run in someone else's cloud. Kartā turns the ChatGPT plan you already pay for into a private coding agent on a server you own. ~$15/mo. Your code never leaves your disk. Every turn sandboxed.
why bother
The hosted tools — Devin, zo.computer, Cursor's background agents — are real products. Pick one if you want zero setup and someone else's uptime to manage. Pick Kartā if any of these matter to you.
01 · privacy
Tailscale-only. No public IP. Every secret encrypted at rest. Your repos live on a server you own — never a vendor's database, never a sandbox in someone else's data center.
02 · the unlock
Kartā runs on the ChatGPT plan you already pay for — Plus, Pro, Business, or Enterprise. No second AI subscription. The cockpit + the host: ~$15/mo on Hetzner, $0 → ~$60/mo on AWS Free Tier (≈3 months free), or €0 on a Raspberry Pi.
03 · ownership
Walk away with your data any time — one command stops everything, your sessions and audit log stay on your disk. No SaaS contract, no hosted tenant to migrate out of, no "we're sunsetting that feature" notice on a Tuesday.
what it does
Eight things that earn their pixels. Each one is the reason something else didn't quite fit your shape.
Every command, every file change, every reasoning step — streamed live as the agent writes them. Vibe coding, with receipts.
Click Open PR. The diff, bot reviews, and checks appear in the same page. Click Squash and merge. Done — no tab-hopping to GitHub mid-session.
Deep-link any session, open it on your phone — the worklog picks up where the agent left off. Optional Telegram bridge: wire up your own bot and get pinged when long turns finish.
With the optional Telegram bridge wired up, send a voice note — it becomes a turn. Long prompts are faster spoken. Transcription happens on your server — your audio never leaves your tailnet.
"Every Monday, open PRs for any dependency updates." Real agent sessions on cron, with overlap handling baked in. Telegram pings when each finishes.
When the agent needs to test the UI it just built, a real Linux desktop sits inside the cockpit. A bundled browser-automation CLI lets it verify its own work, headlessly. No competitor ships this.
Every prompt-to-completion cycle runs in a fresh gVisor-isolated container. A user-space kernel intercepts every syscall; when the turn ends, the container is destroyed. Always on, no toggle. Runaway scripts and malicious tool calls can't escape into your real workspace.
Full-text across every title, every message, every turn — straight from the sidebar or ⌘K palette. Hit s to focus, type, see snippets with the match in context. It all runs on your box — your transcripts never leave the tailnet.
how it works · 30-sec version
The only public-facing things are GitHub and OpenAI — both of which you already use. Kartā never serves a public IP. You reach the cockpit over your tailnet, which IS the firewall.
three themes, one keystroke
Three carefully-paired dark palettes. Switch via the dots in the sidebar. Saved per device.
vs the alternatives
Different bets on who manages what. The honest, qualitative cut — pricing and feature parity in this space shifts monthly.
your ChatGPT plan + ~$15/mo
$500+/mo · hosted
subscription · browser sandbox
getting started
Plan to spend ~10 min collecting credentials in your browser, ~20 min watching the provisioning ladder run. The TUI walks you through every prereq in-line — no separate checklist. See the full install guide →
The agent runs on someone else's GPU. Everything else can run on a server you own.
— that's the whole bet