> ## Documentation Index
> Fetch the complete documentation index at: https://docs.zeplo.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# MCP Server

> Connect AI assistants to your Zeplo workspace using the Model Context Protocol. Manage campaigns, contacts, conversations, and more directly from Claude, Cursor, or any MCP-compatible tool.

# MCP Server

The Zeplo MCP Server exposes your WhatsApp marketing workspace as a set of **43 tools** and **3 resources** that any MCP-compatible AI assistant can use directly — no HTTP calls, no SDK boilerplate.

## What is MCP?

The [Model Context Protocol](https://modelcontextprotocol.io/) is an open standard that lets AI assistants interact with external systems through a structured protocol. Instead of writing API calls, you describe what you want in natural language and the assistant uses the right tool automatically.

## What You Can Do

> **Manage Contacts**
>
> Search, create, update, and delete contacts across all your channels.

> **Run Campaigns**
>
> Create campaigns, check delivery stats, and track performance metrics.

> **Send Messages**
>
> Send text and template messages directly through conversations.

> **View Analytics**
>
> Pull dashboard metrics, message delivery stats, and campaign comparisons.

> **Control Automations**
>
> List, toggle, and trigger automation flows for contacts.

> **Train Chatbots**
>
> Add training data and monitor your AI chatbot configurations.

## Tool Categories

| Category | Tools | Description |
| - | - | - |
| **System** | 3 | Health checks, plans, subscription status |
| **Channels** | 3 | List, inspect, and diagnose WhatsApp channels |
| **Contacts** | 6 | Full CRUD with search and group filtering |
| **Templates** | 3 | List, inspect, and sync WhatsApp templates |
| **Campaigns** | 6 | Campaign lifecycle with analytics |
| **Conversations** | 5 | List, close, assign, and read messages |
| **Messages** | 4 | Send text/template messages, view stats |
| **Analytics** | 4 | Dashboard, message, campaign, and export data |
| **Automations** | 5 | List, toggle, trigger, and view executions |
| **Chatbots** | 4 | List, configure, train, and inspect training data |

## Resources

The server also exposes read-only resources that AI assistants can query:

| Resource | URI | Description |
| - | - | - |
| Schema | `zeplo://schema` | Database schema metadata with all table definitions |
| Plans | `zeplo://plans` | Available subscription plans and features |
| API Endpoints | `zeplo://api-endpoints` | Complete REST API endpoint catalog |

## Requirements

* **Node.js** 18 or later
* **PostgreSQL** database (same one used by your Zeplo instance)
* An MCP-compatible client (Claude Code, Cursor, Windsurf, etc.)

## Next Steps

> **[Setup & Configuration](https://docs.zeplo.ai/docs/mcp-server/setup.md)**
>
> Install and configure the MCP server for your AI assistant.

> **[Tool Reference](https://docs.zeplo.ai/docs/mcp-server/system-tools.md)**
>
> Browse all 43 tools with parameters and examples.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.