Connect an AI agent (MCP)

Sitova exposes a remote Model Context Protocol (MCP) server so AI agents can create pages, edit content, publish, manage articles, media, menus and more, on your behalf. Every action runs through the same plan and permission checks as the dashboard, so an agent can never exceed your plan's limits.

Agent access is available on every plan and is metered: each tool call counts against a monthly Agent (MCP) call budget (Free 200, Pro 2000, Business 10000). Read and write calls both count as one.

Point your MCP client at the server URL for your account, which is https://mcp.<your-root-domain>/mcp. In Cursor, add it to your MCP settings; in Claude or ChatGPT, add it as a custom connector.

The first time you connect, the agent runs a standard OAuth sign-in: a browser window opens on the MCP domain, you sign in with your Sitova credentials and approve the requested permissions (read, write, publish). The agent then receives a token and your tools appear.

You approve exactly which scopes an agent gets: read-only, write, or publish. Publishing still requires an Owner or Admin role and a verified email, and premium blocks are gated by your plan, just like in the dashboard.

Destructive actions such as deleting a page or article are clearly marked so your agent asks before running them.

See every agent you have connected, and revoke any of them instantly, from Settings then Connected agents in your dashboard. Revoking stops the agent from getting new access immediately.