deploymill turns “spin me up a Python API at api.example.com” into a live URL — through one tool call from Claude or any AI agent. No dashboards. No CLI. No yak shaving.
deploymill is an MCP server that wires GitHub and Dokploy into a single tool surface any AI agent can drive.
start_project creates the repo, scaffolds the code, wires CI, deploys, and attaches a domain — in a single call.
Every pull request gets a real, HTTPS preview deployment at a pretty hostname. Reviewers click; nothing to install.
Declare named Docker volumes in .deploymill/project.json. Your SQLite, uploads and caches survive every redeploy.
Opt-in rollback records every deploy as a tagged image. Revert in seconds — no rebuild, no downtime.
Each workspace is isolated to its own Dokploy project, gated by Better Auth. Bring any MCP client — Claude, your own agent, anything.
Static, Node, and Python templates ship out of the box — each one a Dockerfile + entry file, dropped into your repo on day zero.
Domains, mounts, and rollback all live in .deploymill/project.json. Edit, commit, reconcile_project. Done.
Merge or replace env vars per app while preserving comments and Dokploy template refs. Values are redacted in responses.
Sign up with email and password. We mint OAuth tokens for your MCP clients — no shared API keys.
Point Claude (or any MCP client) at https://deploymill.com/mcp. Dynamic Client Registration + PKCE — it just works.
“Ship me a Node API.” “Roll back the last deploy.” “Add a volume at /data.” deploymill turns intent into infrastructure.