--- title: "Deploy tlbx" description: "Browser terminal and coding-agent workstation with a persistent volume" category: "AI/ML" url: https://railway.com/deploy/tlbx --- # Deploy tlbx Browser terminal and coding-agent workstation with a persistent volume **[Deploy tlbx on Railway](https://railway.com/template/tlbx)** Machine-readable deploy manifest (JSON, validated by TemplateCI): https://railway.com/deploy/tlbx/manifest.json - **Creator:** Paul Zarudnev's Projects - **Category:** AI/ML ## Template content ### caddy - **Source:** https://github.com/RockinPaul/tlbx_railway_template - **Health check:** /up - **Public domain:** Yes ### tlbx - **Source:** https://github.com/RockinPaul/tlbx_railway_template ## Documentation # Deploy and Host tlbx on Railway [tlbx](https://github.com/tlbx-ai/tlbx) is a self-hosted terminal multiplexer you reach from a browser: persistent shells, a file and Git view, and dedicated sessions for coding agents like Claude Code, Codex, Gemini CLI and OpenCode. This template runs it as a workstation you can open from a laptop, a tablet or a phone. ## About Hosting tlbx tlbx expects to run on a machine you own, and it has no HTTP mode at all — it serves HTTPS with a certificate it generates for itself. Railway's edge speaks plain HTTP to a container, so this template puts a small gateway in front: the gateway is the only service reachable from the internet, it answers the platform health check itself, and it forwards to tlbx over HTTPS on the private network. Getting the forwarded headers wrong here produces a deployment that looks perfect and does not work, so the gateway configuration is the substance of this template. The login password is generated for you rather than typed into a deploy form, and a volume keeps your home directory, your repositories and your installed toolchains across redeploys. ## Common Use Cases - Keep a long-running coding agent working in a box you can check on from your phone. - Use a real terminal and a Git view from a tablet, or from a machine you would rather not install a toolchain on. - Give a project a persistent home with its own toolchain, separate from your laptop. ## Dependencies for tlbx Hosting - A volume for the workstation, which the template creates. - Nothing else. There is no database, no external service, and no account to register. ### Deployment Dependencies - [tlbx](https://github.com/tlbx-ai/tlbx) — the upstream project (AGPL-3.0-or-later, with commercial licensing available). - [Caddy](https://caddyserver.com) — the gateway (Apache-2.0). - [mise](https://mise.jdx.dev) — installs language toolchains onto the volume (MIT). ### Implementation Details Two services build from this template's repository, each from its own directory: - **caddy** — the only public service. It answers the platform health check on its own, then forwards everything to tlbx over HTTPS with verification skipped, because tlbx's certificate is self-signed and the hop never leaves Railway's private network. It passes the client's Host through verbatim: tlbx compares the browser's Origin against that Host on scheme, host and port, so a rewritten Host makes every WebSocket the interface opens fail while the page still loads. - **tlbx** — the server and your shells, built on a Node base image because the coding agents people run inside tlbx are npm packages. The upstream project publishes no container image, so the template installs the signed release binary and checks it against a pinned digest. Your home directory, npm global prefix and mise data all live on the volume. The password is generated with a Railway secret, so there is nothing to fill in. Read it from the tlbx service's variables, then open the gateway's URL and sign in. Installing a toolchain with `mise use -g python@3.13` puts it on the volume, so it survives redeploys; the image itself ships only Node. Three limits are worth knowing before you deploy. Running sessions end when the service redeploys, because a redeploy replaces the host — files on the volume survive, shells do not. The upstream app preview feature cannot work here, because it serves previews from a second port and tells the browser to fetch them from a URL Railway's edge does not answer. And Railway containers are not privileged, so nothing inside can build or run containers. ## Why Deploy tlbx on Railway? Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it. By deploying tlbx on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway. ## Similar templates - [Chat Chat](https://railway.com/deploy/-WWW5r) — Chat Chat, your own unified chat and search to AI platform. - [stella](https://railway.com/deploy/stella) — Self-host stella with web, API, Postgres, Redis, and object storage. - [Hermes Agent | OpenClaw Alternative with Dashboard](https://railway.com/deploy/hermes-agent-or-openclaw-alternative-wit) — Self-Hosted Hermes AI Agent for Telegram, Discord & Slack Open this page in a browser: https://railway.com/deploy/tlbx