---
title: "Deploy openocta"
description: "OpenOcta is an AI hub that combines gateway, agent runtime, channels, cron"
category: "AI/ML"
url: https://railway.com/deploy/openocta
---

# Deploy openocta

OpenOcta is an AI hub that combines gateway, agent runtime, channels, cron

**[Deploy openocta on Railway](https://railway.com/template/openocta)**

- **Creator:** kmaki's Projects
- **Category:** AI/ML

## Template content

### openocta-image

- **Image:** xiaosong233/openocta-railway:latest
- **Public domain:** Yes

## Documentation

# Deploy and Host OpenOcta on Railway

## About Hosting

[OpenOcta](https://github.com/openocta/openocta) is an open-source, enterprise-oriented operations agent platform. It packages a Gateway (HTTP + WebSocket), Agent runtime, channels, cron, and embedded Control UI into a single Go backend binary. For Railway deployment, this template uses a prebuilt image source (`docker.io/xiaosong233/openocta-railway:latest`) to avoid upstream build drift and keep runtime behavior stable.

The service listens on port `18900` by default (`openocta gateway run --port 18900`). This template maps Railway routing to that port with `PORT=18900`, while preserving upstream command behavior.

## Why Deploy

- **Single-service architecture** — Gateway and embedded frontend ship in one deployable unit.
- **Upstream-aligned build** — Uses official repository and Dockerfile, reducing maintenance drift.
- **Ops-focused workflows** — Designed for observability, troubleshooting, automation, and agent-assisted operations.
- **No external database required** — Core deployment runs without mandatory Postgres/Redis dependencies.
- **Simple production entrypoint** — Built-in command starts both HTTP and WebSocket on one port.

## Common Use Cases

- **Internal operations copilots** for incident triage and troubleshooting workflows.
- **Unified control plane** connecting logs, metrics, CMDB, and ticketing integrations.
- **Webhook-driven automations** via Gateway hook endpoints.
- **Team-accessible control UI** for operating and monitoring agent workflows.
- **Protocol-compatible migration path** from OpenClaw-style Gateway behavior.

## Dependencies for Deployment

### Deployment Dependencies

| Dependency | Details |
|---|---|
| **Docker Image** | `docker.io/xiaosong233/openocta-railway:latest` |
| **Source Type** | Image deployment (not repo build) |
| **Container Port** | `18900` |
| **Public Domain** | Railway-provided service domain |
| **Required Env Vars** | `PORT=18900`, `OPENOCTA_RUN_MODE=service` |
| **Optional Env Vars** | `OPENOCTA_STATE_DIR`, `OPENOCTA_SKIP_CHANNELS`, `OPENOCTA_SKIP_CRON`, `OPENOCTA_SKIP_PROVIDERS`, `ANTHROPIC_API_KEY` |
| **Persistent Volume** | Optional; mount path recommended: `/root/.openocta` (for durable local state/config) |
| **Database** | None required |
| **Start Command Override** | Not required (use upstream `ENTRYPOINT/CMD`) |

## 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/openocta