AI Agents experimental 1 service 2 GiB RAM 10 GB disk

Kilo (headless)

Kilo in server mode - a persistent remote coding agent your IDE extension and CLI clients attach to.

One-click deploy, from $13/mo on a Miget plan.

Kilo is an open-source AI coding agent; kilo serve is its headless mode - an HTTP+SSE API that Kilo clients attach to. Run it on a platform and the agent outlives your editor: the IDE extension, kilo attach from a terminal, or the Cloud Agents web UI (via Remote Connections) all connect to the same persistent brain.

This template runs server mode behind HTTP basic auth with a 10 GB /workspace volume that survives redeploys. Clone your repos into the workspace once; every client that attaches sees the same state.

Two variables: KILO_SERVER_PASSWORD for the API and ANTHROPIC_API_KEY (or configure another provider through Kilo’s config in the workspace). Prefer a terminal over an API? The agent-box template ships the Kilo CLI alongside Claude Code and opencode in a browser terminal.

#what you get

  • kilo serve: HTTP+SSE API for IDE, CLI, and web clients
  • Persistent 10 GB /workspace across redeploys
  • Attach from the IDE extension, kilo attach, or Cloud Agents UI
  • HTTP basic auth on the API
  • Provider-flexible: Anthropic by default, others via Kilo config
  • Single container, official open-source agent

#topology

ServiceRolePublic
kilokilo serve - HTTP+SSE API (:5000)yes (basic auth)

#miget sizing

// this stack needs

2 GiB RAM · 10 GB disk · 1 service

2 GiB fits the agent runtime and typical repo tooling. Big monorepos or heavyweight builds in the workspace want the next plan up.

Hobby - recommended fit

$13/mo

1 vCPU · 2 GiB · 50 GiB disk

Headroom for your own apps: 2 GiB at $19/mo

Professional - production

$22/mo

1 vCPU · 2 GiB · 10 GiB disk

Dedicated resources, production SLOs - plan details

One Miget plan is a fixed pool of compute - the whole stack (managed databases included) deploys inside it, and anything left over runs your other apps. No per-service or per-seat math.

#vs. other PaaS

Estimated monthly cost of running this exact stack (2 GiB RAM, 10 GB disk, 1 container) elsewhere, from published June 2026 rates.

PlatformEst. monthlyNotes
Miget $13 flat compose stacks first-class: one deploy, dedicated vCPU, managed Postgres/Valkey, volumes and TLS all included in the plan
Heroku ~$100 no volumes; nothing between 1 GB ($50) and 2.5 GB ($250) dynos - 2 GB containers cost far more than shown
DO App Platform ~$29 no persistent volumes - stateful containers need managed DBs/Spaces (base $5 Spaces included here)
Render ~$28 per-service instances (0.5 GB $7, 2 GB $25) - every container is its own paid service
Railway ~$22 usage-based ($10/GB RAM-mo); vCPU billed separately at $20/vCPU-mo on top
Fly.io ~$13 cheapest sticker price - but burstable shared CPUs (1/16 core; dedicated vCPUs cost ~2-3×), no compose deploys (one app per container, manual wiring), managed DBs billed extra

Estimates assume RAM fully allocated at published on-demand rates - and sticker price isn't the whole comparison: the cheaper rows buy burstable shared CPUs, per-service wiring instead of a compose deploy, and managed databases billed separately. Heroku and DO App Platform have no persistent volumes at all - stateful stacks like this one need workarounds there.

#deploy it

On Miget

  1. Create a Compose Stack in app.miget.com pointing at the templates repository
  2. Set the stack path to kilo
  3. Set the required variables:
    • KILO_SERVER_PASSWORD, HTTP basic auth on the API
    • ANTHROPIC_API_KEY, model access, billed to you (other providers via Kilo config)
  4. Deploy. Miget layers compose.miget.yaml (RAM, privacy, volumes, managed services) automatically

Locally first?

Every template is portable, vanilla Docker Compose - the Miget overrides are ignored locally:

git clone https://github.com/deployable-sh/stacks
cd miget-compose-templates/kilo
docker compose up -d

Same files, same behavior. The template README covers connection strings and scaling notes.

#faq

What is the difference between Kilo and agent-box?

Interface. This template exposes Kilo’s server API for programmatic and IDE clients; agent-box gives you a browser terminal with the Kilo CLI (plus Claude Code and opencode) inside. Same agent family, API versus terminal.

How do I connect my IDE to it?

Point the Kilo extension (or kilo attach) at https://your-app-domain with KILO_SERVER_PASSWORD as the basic-auth credential. The Cloud Agents web UI connects the same way via Remote Connections.

Why run Kilo remotely instead of locally?

Persistence and reach: long tasks keep running after your editor closes, the workspace state is shared by every client you attach from, and a phone or tablet can check on (or redirect) the agent mid-task.

Why is this marked experimental?

Kilo’s server mode and remote-connection protocol are young and moving fast upstream. It works today; expect the template to track breaking changes. Pin image versions for stability.

Ship Kilo (headless) today

One compose stack, 2 GiB of RAM, from $13/month flat, and it runs on your laptop with the same files.