SendPulse chatbots MCP server
MCP Server for SendPulse Chatbots
An MCP server for SendPulse chatbots: account info, dialogs and sending messages across WhatsApp, Telegram, Instagram and other channels
- 0 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Can send messages to messenger contacts on the bot's behalf
- The x-sp-id and x-sp-secret headers or the OAuth token grant access to all of the account's bots
Meta, the owner of Facebook, Instagram and WhatsApp, is designated an extremist organization in Russia and its activities are banned there.
Install
- Works in
- Services
Manual install
npm install && npm run build && npm startInstalls, builds and starts the local HTTP server on port 3000.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Can send messages to messenger contacts on the bot's behalf
- The x-sp-id and x-sp-secret headers or the OAuth token grant access to all of the account's bots
This is third-party code. Review the repository files before installing.
What it does
The server runs as an Express HTTP service and gives an agent three general tools: SendPulse account info with plan and counters, a list of connected chatbots, and a list of dialogs across all channels with pagination and sorting. A separate universal send_message tool sends a text message to a contact on a chosen channel, supporting WhatsApp, Telegram, Instagram, Messenger, LiveChat and Viber. Authentication is flexible: you can pass an application ID and secret in headers and the server obtains and caches an OAuth token itself, or pass a ready token directly via the Authorization header or the initialize request body.
Who it is for. For people running a SendPulse chatbot across multiple messengers who want to read dialogs and send messages through an agent.
Good fit when
- You want a summary of dialogs across all connected bot channels in one place
- You want to send a message to a contact on a specific channel without opening the SendPulse dashboard
- You want SendPulse account and plan information
Not a fit when
- You do not have a SendPulse account with connected chatbots
- You want to test the server locally with an external AI client: this needs a public tunnel like ngrok, which exposes the server to the internet
Example request
Show recent dialogs in my SendPulse chatbots and send a reply to a contact on WhatsAppMCP
- Transport
- http
- Authentication
- OAuth
Limitations
The project has a single author, is very small, with no ready npm package, and installs from source by building the TypeScript. Only sending a plain text message is available; there is no support for templates, attachments or buttons. The README's ngrok test setup exposes the local server to the internet with no default authorization.
How to disable. Stop the Node.js server process and remove the entry from your MCP client config.
FAQ
Which channels does message sending support?
WhatsApp, Telegram, Instagram, Messenger, LiveChat and Viber, chosen via the channel parameter of the send_message tool.
Can I pass a ready OAuth token instead of an ID and secret?
Yes, a token can be passed directly via the Authorization header or in the initialize request body.
Discussion
No questions or reviews yet.
Related
All SendPulse toolsSendPulse MCP server for WhatsApp and SMS
SendPulse MCP Server
An MCP server for SendPulse WhatsApp bots and SMS: contacts, button messages, flows, triggers and live chat
SendPulse MCP server
SendPulse MCP
SendPulse's official remote MCP server: statistics, campaigns and user data through an agent chat
SendPulse MCP server
sendpulse-mcp
An MCP server for SendPulse: email campaigns, mailing lists, transactional mail via SMTP and web push notifications
ChatbotX MCP server: contacts, conversations, flows, broadcasts and automation in WhatsApp, Instagram and other channels
Official LINE MCP server: an agent sends messages through the Messaging API from your LINE Official Account
Official Chatway MCP server: the agent triages live chat conversations, replies to customers and searches the knowledge base
README in short
The English README describes three general tools and the universal send_message tool with a channel parameter, two authentication methods, installation and building on Node.js and Express, running the local HTTP server, and a test setup exposing it via ngrok for external AI clients like the OpenAI sandbox.