MCP server
Ownkube runs a hosted remote MCP server. Point any MCP-capable client at one URL, sign in once, and your agent can ship deployments, tail activity, and manage environments on your behalf. It uses the same permissions and tenant isolation as your dashboard account, and Ownkube never receives your cloud keys.
MCP (the Model Context Protocol) is an open standard for connecting agents to tools. Because this is a remote server that speaks OAuth, any client with remote-MCP support connects the same way: Claude, other desktop and web assistants, agent frameworks, and your own code via an MCP SDK. The walkthrough below uses Claude as a concrete example; the URL and the flow are identical everywhere else.
The server URL
Section titled “The server URL”https://app.ownkube.io/api/mcpThat is the only value you need. Authentication happens in the browser: there is no token or API key to copy, and nothing secret to paste into a config file.
Connect your coding agent
Section titled “Connect your coding agent”Every client follows the same shape: point it at the server URL, authorize in the browser, pick an organization. The only thing that differs is where you add the server. Pick your client below for the exact step.
Clients that speak HTTP MCP (Claude Code, Cursor, GitHub Copilot, Windsurf) connect to the URL directly. Terminal and desktop clients that speak stdio (Codex, Zed, Claude Desktop) use mcp-remote, a small bridge that connects a local client to the remote endpoint and runs the browser sign in for you. It needs Node and npx on your machine.
-
Install Claude Code
Terminal window npm install -g @anthropic-ai/claude-code -
Add the server
Terminal window claude mcp add --transport http ownkube https://app.ownkube.io/api/mcpStreamable HTTP is what the
--transport httpflag targets. On first use Claude Code registers itself and starts the browser sign in automatically. -
Authenticate
Run
/mcpinside Claude Code and choose ownkube. A browser tab opens for you to sign in and pick an organization, then control returns to your terminal.
-
Open the MCP settings
Download Cursor from cursor.com, then open Settings > MCP.
-
Add the server
Add it to
~/.cursor/mcp.json(or a project-level.cursor/mcp.jsonto scope it to one repo):~/.cursor/mcp.json {"mcpServers": {"ownkube": {"url": "https://app.ownkube.io/api/mcp"}}} -
Authenticate
Toggle the ownkube server on in Settings > MCP. Cursor opens a browser tab to sign in and pick an organization, then shows the server as connected.
-
Enable agent mode
Install VS Code and the GitHub Copilot extension, then enable agent mode.
-
Add the server
Add it to the workspace config at
.vscode/mcp.json:.vscode/mcp.json {"servers": {"ownkube": {"type": "http","url": "https://app.ownkube.io/api/mcp"}}} -
Authenticate
Open the MCP view and start the ownkube server. VS Code opens a browser tab for you to sign in and choose an organization, then marks the server running. Switch Copilot Chat to Agent mode to use the tools.
-
Open the Cascade MCP settings
Download Windsurf from codeium.com/windsurf, then open the Cascade MCP settings.
-
Add the server
~/.codeium/windsurf/mcp_config.json {"mcpServers": {"ownkube": {"serverUrl": "https://app.ownkube.io/api/mcp"}}}Windsurf uses
serverUrlfor a remote MCP server. -
Authenticate
Press refresh in the MCP settings. Windsurf opens a browser tab to sign in and pick an organization, then shows ownkube as available.
-
Install Codex
Terminal window npm install -g @openai/codex -
Add the server
Codex speaks stdio MCP, so bridge it with
mcp-remote:~/.codex/config.toml [mcp_servers.ownkube]command = "npx"args = ["-y", "mcp-remote", "https://app.ownkube.io/api/mcp"] -
Authenticate
The first time Codex starts the ownkube server,
mcp-remoteopens a browser tab for you to sign in and choose an organization. After that it reuses the saved token.
-
Open settings
Download Zed from zed.dev, then open
settings.json. -
Add the server
Zed calls stdio MCP context servers, so bridge it with
mcp-remote:settings.json {"context_servers": {"ownkube": {"source": "custom","command": "npx","args": ["-y", "mcp-remote", "https://app.ownkube.io/api/mcp"]}}} -
Authenticate
Save the settings and open the agent panel.
mcp-remoteopens a browser tab for you to sign in and pick an organization, then Zed lists the ownkube tools.
-
Install Claude Desktop
Download Claude for macOS or Windows from claude.ai/download.
-
Add the server
Bridge it with
mcp-remote:claude_desktop_config.json {"mcpServers": {"ownkube": {"command": "npx","args": ["-y", "mcp-remote", "https://app.ownkube.io/api/mcp"]}}}On paid plans you can also add Ownkube as a custom connector by URL from Settings > Connectors.
-
Authenticate
Restart Claude Desktop after saving. On first use
mcp-remoteopens a browser tab for you to sign in and pick an organization, then the ownkube tools appear in the chat.
Once connected, ask for what you need, for example “list my Ownkube deployments” or “roll back the api deployment to the previous revision”. Your agent picks the right tool and calls it.
What your agent can do
Section titled “What your agent can do”The MCP server mirrors the Ownkube API surface. Your agent can read and act across the resources you own.
Inspect (read):
| Area | What the agent can see |
|---|---|
| Deployments | Deployments, status, revision history, and live activity |
| Environments | Environments and their variables |
| Registries | Connected container registries |
| Organizations | The organizations your account belongs to, and your user profile |
Act (write):
| Area | What the agent can do |
|---|---|
| Deployments | Create and delete deployments, update config, upgrade platform versions, roll back, and promote one deployment’s config onto another |
| Environments | Create and update environments and set their variables |
| Jobs | Trigger, cancel, and list job runs |
Own cloud (Advanced). If you run in your own AWS account, the same connector also exposes these. On Ownkube-hosted compute they simply don’t apply.
| Area | Read | Write |
|---|---|---|
| Clusters | Clusters, their status, and lifecycle state | Create, cancel, and destroy clusters |
| Cloud accounts | Connected AWS accounts and their onboarding state | Connect and reconnect AWS accounts |
Write actions change real infrastructure. Tools that create, upgrade, roll back, or destroy are described so the agent confirms with you before running them, but you are in control: treat an agent with this connector the way you would treat a teammate with dashboard access.
Security model
Section titled “Security model”- OAuth, not keys. The connection is authorized with OAuth in your browser. There is no long-lived key to leak, and you can revoke the connector from Claude at any time.
- Scoped to your account. The agent sees exactly what your account can see, in the one organization you approved. The same tenant isolation that guards the dashboard guards every tool call.
- Your cloud stays yours. Ownkube never receives your AWS or GCP keys through this server, and the agent cannot read them out. Cloud access is the scoped, least-privilege role you granted during AWS onboarding.
- Auditable. Every action runs through the same API as the dashboard and CLI, so writes are recorded the same way.
Multiple organizations
Section titled “Multiple organizations”The connection is pinned to the organization you picked at approval time. To act on a different organization, add the connector a second time and choose the other organization when prompted, or remove and re-add it. Read-only questions about which organizations you belong to work regardless of the pinned one.