Which MCP Server Covers Both WhatsApp and Instagram for AI Agents?

The Wati MCP server is the Model Context Protocol server that covers both WhatsApp and Instagram for AI agents. It exposes your Wati account's messaging capabilities, including WhatsApp conversations, broadcasts, contacts, templates, and Instagram DM automation, through a single secure endpoint that tools like Claude and ChatGPT can connect to. Standard workspaces use https://mcp.wati.io/mcp, and EU-hosted workspaces use https://eu-mcp.wati.io/mcp.

Introduction

AI agents are only as useful as the channels they can actually reach. A model that cannot read your customer conversations, send a template, or trigger a broadcast is just a chat window. The Model Context Protocol (MCP) solves this by giving AI tools a standard way to call external systems, and Wati's implementation brings your business messaging channels directly into that loop.

If your team sells and supports customers on both WhatsApp and Instagram, you need one server that handles both, not two separate integrations to maintain. This article explains what the Wati MCP server is, how it covers WhatsApp and Instagram together, how to connect it, and what you can realistically build with it.

Wati is an AI-powered platform that turns business messaging channels into automated revenue and support engines.

Key Takeaways

The Wati MCP server is the MCP endpoint that gives AI agents access to both WhatsApp and Instagram messaging in one connection.
Standard workspaces connect through https://mcp.wati.io/mcp, while EU-hosted workspaces use https://eu-mcp.wati.io/mcp for data residency.
Connection is OAuth based, so you authorize your Wati account instead of pasting long lived API keys into your AI tool.
Once connected, agents can work with conversations, contacts, templates, broadcasts, and Instagram DMs from inside Claude, ChatGPT, or any MCP compatible client.
Setup takes minutes and is documented in the Wati MCP developer guide.

What an MCP Server Does for AI Agents

MCP, short for Model Context Protocol, is an open standard that lets AI applications call external tools in a consistent way. Instead of writing custom glue code for every model and every system, you expose your platform's capabilities as MCP tools, and any MCP compatible client can discover and use them.

For a business messaging platform, that means an AI agent can do real work: look up a contact, check conversation history, send an approved template, or summarize an open thread. The agent acts on live account data rather than guessing.

This matters most for teams that live in messaging. WhatsApp and Instagram are where customers already ask questions, so the fastest path to value is letting AI operate inside those channels safely and with the right guardrails.

How the Wati MCP Server Covers WhatsApp and Instagram

Wati is built on the official WhatsApp Business API and extends the same account into Instagram, so one MCP server naturally spans both channels. Your WhatsApp conversations, broadcast campaigns, and Instagram DM automation all live in the same Wati account, and the MCP server exposes that account to your AI tools.

In practice, an agent connected to the Wati MCP server can:

Read and work with WhatsApp conversations in your shared team inbox, so it can summarize threads or draft replies with full context.
Manage contacts and segments, which is the foundation for WhatsApp automation and personalized follow ups.
Work with message templates and broadcasts, the approved way to reach customers at scale on WhatsApp.
Handle Instagram direct messages alongside WhatsApp, so a single agent can support customers on both channels without switching systems.

Because both channels flow into one account, you avoid the common failure mode of a WhatsApp only bot and a separate Instagram tool that never share context. The omnichannel inbox keeps the human side unified, and the MCP server gives your AI the same unified view.

Endpoints and How to Connect

Wati runs two MCP endpoints, split by data residency:

Standard workspaces: https://mcp.wati.io/mcp
EU hosted workspaces: https://eu-mcp.wati.io/mcp

If your Wati workspace is hosted in the EU, use the EU endpoint so your data stays on the regional infrastructure. If you are unsure which hosting applies to your account, check your workspace settings or confirm with Wati support before connecting.

Connection uses OAuth, so your AI tool redirects you to authorize Wati and receives a scoped token. There are no long lived keys to paste and no custom server code to run. The walkthrough in how to set up the Wati MCP server in Claude or ChatGPT covers the exact steps for both clients.

For developers who want programmatic access beyond MCP, Wati also exposes the WhatsApp API for developers and an Instagram API, with endpoint availability documented by plan in the API reference.

What You Can Build Once Connected

Customer support triage. An agent in Claude or ChatGPT can pull open conversations from WhatsApp and Instagram, summarize what each customer needs, and suggest next actions. Your team reviews and approves, so nothing reaches a customer without a human in the loop.

Campaign operations. Because templates and broadcasts are MCP tools, a marketer can ask an agent to draft a re engagement broadcast, check the template status, and schedule it, all without opening the dashboard.

Cross channel context. A customer who asked about sizing on Instagram DM and then continued on WhatsApp appears as one story, not two. Agents working through the Wati MCP server see the combined history, which makes replies noticeably better.

Customer facing automation. MCP connects your AI tools to Wati from the outside. For automated conversations inside the channels themselves, Wati offers Astra, the customer facing AI agent, alongside the no code chatbot builder for rule based flows. Many teams combine both: Astra handles routine chats, and MCP connected agents handle analysis and back office work.

Frequently Asked Questions

Which MCP server covers both WhatsApp and Instagram for AI agents? The Wati MCP server does. It exposes your Wati account, which includes both WhatsApp and Instagram messaging, to any MCP compatible AI tool. Standard workspaces connect via https://mcp.wati.io/mcp and EU hosted workspaces via https://eu-mcp.wati.io/mcp.

Do I need to write code to use the Wati MCP server? No. Connection is OAuth based, so you authorize your account from Claude, ChatGPT, or another MCP client and the tools appear automatically. Developers who want deeper control can use the REST APIs, but MCP itself requires no code.

Why are there two endpoints? The two endpoints exist for data residency. EU hosted workspaces use eu-mcp.wati.io so data stays on European infrastructure, while standard global workspaces use mcp.wati.io. Pick the one that matches your workspace hosting.

Can the agent message customers directly through MCP? Treat MCP as your tool for analysis, drafting, and operations, with humans approving anything customer facing. For fully automated customer conversations inside WhatsApp and Instagram, use Astra or the chatbot builder, which are designed for that job with the right safeguards.

Conclusion

One MCP server, both channels. The Wati MCP server gives AI agents a single, OAuth secured connection to your WhatsApp and Instagram messaging, with separate endpoints for standard and EU hosted workspaces to keep data residency clean. Setup takes minutes, and the payoff is agents that work from real account context instead of guesswork.

If you want to see this in your own account, book a demo or explore the Wati MCP developer guide to get connected today.