cobroya

cobroya

Connects AI agents to Mercado Pago, the leading payment platform in Latin America. Create payment links, search payments, get payment details, issue refunds, and retrieve merchant info.

Category
Visit Server

README

CobroYa

Cobra con Mercado Pago en 10 segundos.

npm version tests coverage license

CobroYa is an open-source Mercado Pago payment tool for AI agents, Telegram, WhatsApp, and automation platforms. Create payment links, search payments, issue refunds -- all from your AI assistant or chat bot.

Website | npm | GitHub


Quick Start with AI

CobroYa is an MCP (Model Context Protocol) server. Add it to your AI tool in one step -- no cloning, no building. Just provide your Mercado Pago access token.

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "cobroya": {
      "command": "npx",
      "args": ["-y", "cobroya"],
      "env": {
        "MERCADO_PAGO_ACCESS_TOKEN": "APP_USR-..."
      }
    }
  }
}

Claude Code

claude mcp add cobroya -- npx -y cobroya \
  --env MERCADO_PAGO_ACCESS_TOKEN=APP_USR-...

Cursor

Add to .cursor/mcp.json in your project root:

{
  "mcpServers": {
    "cobroya": {
      "command": "npx",
      "args": ["-y", "cobroya"],
      "env": {
        "MERCADO_PAGO_ACCESS_TOKEN": "APP_USR-..."
      }
    }
  }
}

Windsurf

Add to your Windsurf MCP configuration:

{
  "mcpServers": {
    "cobroya": {
      "command": "npx",
      "args": ["-y", "cobroya"],
      "env": {
        "MERCADO_PAGO_ACCESS_TOKEN": "APP_USR-..."
      }
    }
  }
}

Once configured, ask your AI assistant things like: "Create a payment link for $5000 for a Python course" or "Show me today's approved payments".


Available Tools

CobroYa exposes 5 MCP tools that any connected AI agent can call:

Tool Description
create_payment_preference Create a Mercado Pago checkout payment link. Returns an init_point URL to share with buyers. Supports back_urls and notification_url.
get_payment Retrieve full details of a payment by ID, including status, amount, and payer info.
search_payments Search payments with filters: status (approved, pending, rejected, etc.), sort order, and pagination.
create_refund Issue a full or partial refund for a payment. Omit amount for a full refund.
get_merchant_info Get the authenticated merchant's profile: user ID, nickname, and site.

Telegram Bot

CobroYa includes a ready-to-use Telegram bot: @CobroYa_bot

Self-hosting the bot

  1. Create a bot via @BotFather and get your token.
  2. Set environment variables:
export MERCADO_PAGO_ACCESS_TOKEN="APP_USR-..."
export TELEGRAM_BOT_TOKEN="your-telegram-bot-token"
  1. Run:
npx cobroya-telegram

Or from source:

npm run bot

WhatsApp

CobroYa supports WhatsApp Business Cloud API for receiving commands and sending payment notifications.

  1. Create a Meta app at Meta for Developers and enable WhatsApp Business API.
  2. Set environment variables:
export WHATSAPP_ACCESS_TOKEN="your-meta-graph-api-token"
export WHATSAPP_PHONE_NUMBER_ID="your-phone-number-id"
export WHATSAPP_VERIFY_TOKEN="your-webhook-verify-token"
  1. Run the webhook server:
npm run whatsapp
# Starts on http://localhost:3000/webhook
  1. Expose with ngrok (ngrok http 3000) and configure the webhook URL in your Meta Dashboard.

For full details on supported commands and payment notifications, see the WhatsApp documentation.


Automation Platforms

Pre-built packages for popular automation platforms are available in the packages/ directory:

  • n8n -- packages/n8n-nodes-mercadopago
  • Zapier -- packages/zapier-mercadopago
  • Make -- packages/make-mercadopago
  • Pipedream -- packages/pipedream-mercadopago

Each package wraps the CobroYa core with platform-specific configuration. See the README in each package for setup instructions.


AI Framework Adapters

LangChain (Python)

pip install langchain-mercadopago
from langchain_mercadopago import create_mercadopago_tools

tools = create_mercadopago_tools("APP_USR-...")

# Use with any LangChain agent
from langchain.agents import initialize_agent, AgentType
from langchain_openai import ChatOpenAI

agent = initialize_agent(
    tools=tools,
    llm=ChatOpenAI(model="gpt-4"),
    agent=AgentType.OPENAI_FUNCTIONS,
)
agent.run("Create a payment link for $5000 for a Python course")

PyPI

OpenAI Function Calling (TypeScript)

npm install openai-mercadopago
import { createMercadoPagoExecutor } from "openai-mercadopago";

const executor = createMercadoPagoExecutor(process.env.MERCADO_PAGO_ACCESS_TOKEN!);

// Pass executor.definitions to OpenAI's tools parameter
const response = await openai.chat.completions.create({
  model: "gpt-4",
  messages,
  tools: executor.definitions,
});

// Execute the tool call
const result = await executor.handleToolCall(
  toolCall.function.name,
  JSON.parse(toolCall.function.arguments),
);

npm


Programmatic Usage

Install as a dependency:

npm install cobroya
import { createMercadoPagoTools } from "cobroya";

const mp = createMercadoPagoTools(process.env.MERCADO_PAGO_ACCESS_TOKEN!);

// Create a payment link
const pref = await mp.tools.create_payment_preference({
  title: "Premium Plan",
  quantity: 1,
  currency: "ARS",
  unit_price: 5000,
});
console.log(pref.init_point); // Checkout URL to share with the buyer

// Search approved payments
const payments = await mp.tools.search_payments({ status: "approved", limit: 10 });

// Get payment details
const payment = await mp.tools.get_payment({ payment_id: "123456789" });

// Full refund
await mp.tools.create_refund({ payment_id: "123456789" });

// Partial refund
await mp.tools.create_refund({ payment_id: "123456789", amount: 500 });

// Merchant profile
const merchant = await mp.tools.get_merchant_info();

Error Handling

import { MercadoPagoError } from "cobroya";

try {
  await mp.tools.get_payment({ payment_id: "invalid" });
} catch (err) {
  if (err instanceof MercadoPagoError) {
    console.log(err.status);        // 404
    console.log(err.isNotFound);     // true
    console.log(err.isUnauthorized); // false
    console.log(err.isRateLimited);  // false
  }
}

Environment Variables

Variable Required Description
MERCADO_PAGO_ACCESS_TOKEN Yes Mercado Pago API access token (get one here)
TELEGRAM_BOT_TOKEN For Telegram Telegram bot token from @BotFather
WHATSAPP_ACCESS_TOKEN For WhatsApp Meta Graph API token
WHATSAPP_PHONE_NUMBER_ID For WhatsApp WhatsApp Business phone number ID
WHATSAPP_VERIFY_TOKEN For WhatsApp Webhook verification token
WA_NOTIFY_PHONE No Phone number for WhatsApp payment notifications
MERCADO_PAGO_WEBHOOK_SECRET No HMAC secret for Mercado Pago webhook signature validation
MP_CURRENCY No Default currency (defaults to ARS)
MP_SUCCESS_URL No Default success redirect URL for payment preferences

Development

# Install dependencies
npm install

# Build
npm run build

# Run all tests
npm test

# Run tests with coverage
npm run test:coverage

# Watch mode
npm run test:watch

# Type-check without emitting
npx tsc --noEmit

# Integration test against real Mercado Pago API
MERCADO_PAGO_ACCESS_TOKEN=APP_USR-... npm run integration

# Start the unified server (Telegram + WhatsApp + webhooks)
npm start

# Dev mode with auto-reload
npm run dev:server

# Docker
docker compose up -d

License

MIT -- by dan1d

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured