Deploy deputies-monolith
Deputies, a control plane for delegated engineering agents.
Just deployed
/var/lib/postgresql/data
Just deployed
control-plane
Just deployed
artifacts
Bucket
Just deployed
Deploy and Host deputies-monolith on Railway
Deputies is a control plane for delegating engineering work to background AI agents.
It gives each task a durable session with queued prompts, live progress, diagnostics, artifacts, and integrations for Slack, GitHub, and webhooks, so teams can run, monitor, and resume agent work across sandboxed environments.
About Hosting deputies-monolith
Hosting deputies-monolith runs the Deputies API, worker loop, event stream, integrations, and Flue runner together in one service.
The template provisions and configures Postgres for durable product state, plus object storage for artifact blobs. It boots with fake runner and sandbox defaults so the UI, auth, setup checks, database, and artifact plumbing work immediately.
To run real agent jobs, configure model credentials, set RUNNER=flue, and choose a sandbox provider such as Daytona.
Common Use Cases
- Delegate engineering tasks to background AI agents and monitor progress from the web UI.
- Trigger and manage agent workflows from Slack, GitHub issues, or webhooks.
- Run sandboxed coding, debugging, research, and artifact-producing jobs that can be resumed or reviewed later.
Dependencies for deputies-monolith Hosting
- Included Postgres database for durable product, queue, artifact metadata, and Flue state.
- Included object storage bucket for artifact blobs.
- Optional model credentials for real agent work, such as Anthropic, OpenAI, or OpenAI Codex auth.
- Optional sandbox provider credentials, typically Daytona, plus a compatible sandbox image.
Deployment Dependencies
- Flue: https://github.com/withastro/flue
- Daytona: https://www.daytona.io/
Implementation Details
The template uses static session auth by default. After deployment, open the Railway project, select the deputies-monolith service, then check the Variables tab for the generated AUTH_STATIC_PASSWORD. Log in with AUTH_STATIC_USERNAME and AUTH_STATIC_PASSWORD.
The template already wires Postgres and object storage. For real agent work, set the runner, model credentials, and sandbox provider:
RUNNER=flue
FLUE_MODEL=
# Choose one model auth path
ANTHROPIC_API_KEY=
OPENAI_API_KEY=
FLUE_OPENAI_CODEX_AUTH_BASE64=
# Recommended sandbox provider
SANDBOX_PROVIDER=daytona
DAYTONA_API_KEY=
DAYTONA_IMAGE=
# For browser-facing deployments, also configure:
WEB_BASE_URL=https://
AUTH_COOKIE_SECURE=true
SERVICE_BASE_DOMAIN=
Why Deploy deputies-monolith 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 deputies-monolith 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.
Template Content
control-plane
ghcr.io/sidpalas/deputies-control-plane:0.2.0artifacts
Bucket
