WhatChimp MCP server for your AI

Plain text through send_text_message only works inside the 24-hour window. Outside it, the agent has to use send_template_message or send_template_with_buttons, which send pre-approved templates and work at any time. get_message_status then tells you whether a send was sent, delivered, read or failed, along with the read time and any failure reason.

Two values, both copied from WhatChimp: your API token and the connected number's Phone Number ID. There is no OAuth handshake and no refresh step to maintain. PopMCP adds both to every call for you, so your prompts never have to mention them, and the token is scoped to a single WhatChimp account.

get_subscriber returns the profile behind a phone number — id, name, email, gender and assigned labels — and list_custom_fields shows the account-wide field definitions before assign_custom_fields writes values to them. add_note leaves an internal note on the chat that the customer never sees. assign_chat_to_team_member hands the conversation to a human agent when the AI should stop.

trigger_bot_flow starts a flow for a subscriber, and get_postback_list returns the postback IDs available on the number so send_template_with_buttons can wire Quick Reply buttons to them. If a contact gets stuck inside a step that collects user input, reset_user_input_flow clears them. The Button tapped trigger fires when a customer taps a quick-reply or postback button.

list_catalogs reads the product catalogs on the account and list_orders returns the orders placed through them — customer details, totals and timestamps — optionally filtered to one catalog. sync_catalog pulls products from Meta into a WhatChimp catalog, and update_order_status moves an order to a new status. Order placed and Order status changed triggers let a workflow react without anyone polling.

The integration covers 30 operations across 6 API modules — 10 reads and 20 writes — plus 11 triggers PopMCP can watch. 22 load by default and the rest stay searchable, so the client's tool list stays readable. Two of the writes, connect_whatsapp_account and get_direct_login_url, are reseller and white-label endpoints that most accounts never call.

No. PopMCP hosts it and gives you a Streamable HTTP MCP URL. There is nothing to deploy, no callback URL to register, and no infrastructure to keep running or patch.

Any client that supports a remote Streamable HTTP MCP server — Claude Desktop, Claude Web, ChatGPT custom connectors, Codex and the Codex CLI, OpenCode and Z.ai's ZCode, as well as editors such as Cursor, Windsurf and Zed. Your client signs in to PopMCP with OAuth; the provider credential never leaves the server.

Loading 30 tool definitions into every conversation burns context before the work starts. PopMCP loads a curated 22-tool set covering the everyday workflows and keeps the rest searchable, so your AI pulls a specialist tool only when a task needs it. You can switch a connection to the full catalog at any time.

A tool is something your AI calls — there are 30 here. A trigger is something PopMCP watches for you: 11 WhatChimp events that can start work on the PopMCP side. Triggers never enter your AI client's context, and none of them ask you to paste a callback URL into WhatChimp.