Deploy with AI

Coming soon

Tell an AI coding agent — Claude Code, Codex, or Cursor — "deploy this to Koo" and watch it handle the rest: use a scoped API token, create the service, deploy it, follow the deployment until it is ready, and report your live URL. It works through the Koo MCP server and the matching skill.

Koo does not run an AI model for you. The agent is your own coding tool, running on your machine, driving Koo through the same API the console uses.

The MCP server and skills are code-complete but not published yet — the koo mcp install command ships with the next CLI release. Deploying a container image works today (see Deploy a container).

How it works

Mint a least-privilege API token — an editor role to deploy, or reader for a read-only agent — scoped to a single account. The token (kc_…) is shown once. koo mcp install can mint one for you if you are logged in with the CLI.
Install the Koo MCP server and skill into your agent:

text

koo mcp install claude-code   # or: codex, cursor

Prompt the agent — "deploy this to Koo". It creates the service, starts a deployment, follows the deployment timeline until the service is ready, and reports the live URL.

To deploy from local source rather than a prebuilt image, the agent has you run koo up in your project directory (the CLI packs and builds it), then watches the result with the same tools.

From there the service is a normal Koo service: the deployment appears in the console, and logs, metrics, and rollback work as usual.

Same API, same guardrails

The agent drives the exact endpoints the console calls — there is no separate deploy path. Your account's roles, plan limits, and audit trail apply to every action the agent takes. Actions are attributed to the token's service account in the audit trail, so you can always see what the agent did, and you can revoke its token at any time.

Until then

Deploying from a container image works today and is the fastest path to a live URL — see Deploy a container.