Deploy OpenDots

OpenDots by CopilotKit: AI coworkers, pages, and private browsing

Deploy OpenDots

Just deployed

Just deployed

/data

Deploy and Host OpenDots by CopilotKit on Railway

Run your own OpenDots workspace with AI coworkers, persistent Spaces and pages, chat, scheduled work, and read-only public-page browsing.

About Hosting OpenDots

This template builds CopilotKit's OpenDots application and its browser service from the public deployment repository. Its Dockerfile pins the upstream application to a specific commit. It generates the owner login token and private browser secret, connects the services over Railway's private network, and stores local application data on a persistent volume. Only the authenticated application receives a public URL.

OpenDots already runs its agents and scheduler in the application, and uses SQLite for local state. This template therefore needs two services and one volume; it does not need Postgres or a separate cron worker. Keep the app at one replica with serverless sleeping disabled.

Dependencies for OpenDots Hosting

You need a Railway account, a CopilotKit Intelligence project, and access to a supported model provider.

Deployment Dependencies

At deployment, provide only your CopilotKit Intelligence API key and OpenAI API key. The template preconfigures OPENAI_MODEL=gpt-6-luna and OPENAI_BASE_URL=https://api.openai.com/v1; no model or URL input is needed. To use another OpenAI-compatible provider, change those two settings to its endpoint and model name and supply its API key.

Open the OpenDots URL and sign in using the generated OWNER_TOKEN from the OpenDots service's Variables tab. Keep that token private. Pages and settings persist on the volume; conversation history is stored by your CopilotKit Intelligence project.

Common Use Cases

  • Organize personal projects using Dots, Spaces, and persistent pages.
  • Chat with AI coworkers using your own model credentials.
  • Run recurring work and capture text from public webpages.

Included and optional features

  • Included: Spaces/pages, configurable Dots, chat, recurring work, and private read-only public-page capture.
  • Optional separate setup: Slack through managed CopilotKit Channels, and voice through a speech provider.
  • Not included: persistent per-Dot computers, interactive browser logins, terminal access, or computer takeover. Upstream implements those with a Docker-socket supervisor that this Railway deployment does not provide. A remote supervisor URL alone is insufficient.

This is a single-owner deployment, not a multi-tenant service. Use a dedicated Railway project and configure volume backups. Public-page browsing does not include a search provider and rejects redirects and private network targets.

Why Deploy OpenDots on Railway?

Railway builds both services from the deployment repository, provides HTTPS for OpenDots, connects Browser over private networking, and attaches persistent storage. You can manage the deployment in one project without a separate Postgres database or cron worker.

Documentation

This independently maintained package pins upstream application code and applies a small GPT-6 Luna tool-calling compatibility adjustment. Review the deployment repository's verification results and limitations when upgrading.


Template Content

More templates in this category

View Template
Chat Chat
Chat Chat, your own unified chat and search to AI platform.

okisdev
116
View Template
stella
Self-host stella with web, API, Postgres, Redis, and object storage.

Jan Kubica
7
View Template
Hermes Agent | OpenClaw Alternative with Dashboard
Self-Hosted Hermes AI Agent for Telegram, Discord & Slack

codestorm
82