Model Context Protocol (MCP)

The Model Context Protocol (MCP) is an open standard that lets AI tools connect to external data and services in a secure, standardized way. The Whazzup MCP server follows the authenticated remote MCP specification and gives AI assistants structured access to your support workspace: finding and retrieving conversations, contacts and knowledge base content, taking actions like replying or changing status, and keeping context about your workspace while they work.

Use the streamable HTTP endpoint for every MCP client. One address serves all regions.

The address is also shown in Whazzup under Settings → MCP server. Clients register automatically – no client ID is required.

1. Your AI tool connects to the Whazzup MCP server.
2. You sign in with your Whazzup account and approve access on a consent screen.
3. The assistant acts as you – with your role, your permissions and your display name. It sees exactly what you see, nothing from other workspaces.
4. The connection stays available so the assistant can read and act whenever you ask.

• Secure access: all data access is authenticated, authorized and limited by row-level security.
• Standardized interface: the same server works with Claude, ChatGPT, Cursor and any other MCP-capable tool.
• Workspace context: the assistant knows your plan, queues and support hours, so answers fit your setup.
• Efficiency: retrieve and act on customer data directly from your AI tool instead of switching tabs.

The MCP server is part of the Unlimited plan, like webhooks and the Data API.