Overview

This MCP (Model Context Protocol) server monitors PancakeSwap on BSC (Binance Smart Chain) and emits structured information when new liquidity pools are created. It continuously watches on-chain factory events, enriches events with token metadata and price/volume estimates, and exposes those results as tools compatible with MCP-aware language models and external integrations.

For developers and ML systems, the server provides real-time alerts (webhooks, sockets), on-demand queries (REST), and historical analytics for newly created pairs. That makes it straightforward to integrate new-pair signals into trading bots, research pipelines, moderation systems, or conversational agents that need current on-chain context.

Features

Real-time detection of newly created PancakeSwap liquidity pools
Token metadata enrichment: symbol, name, decimals, verified contract checks
Price and liquidity heuristics (initial token/BNB amounts, LP token supply)
Configurable alert delivery: webhooks, Telegram, Discord, or custom handlers
MCP-compatible tool endpoints for LLMs and external services
WebSocket and REST APIs for streaming and on-demand access
Persistent storage (SQLite/Postgres) for history and analytics
Docker-ready and environment-configurable for local or production deployment

Installation / Configuration

Prerequisites:

Docker & Docker Compose (recommended) or Node.js (if running locally)
BSC RPC provider URL (public or private node)
Optional: Webhook target, Telegram bot token, Discord webhook URL

Quick start with Docker Compose:

# docker-compose.yml
version: "3.8"
services:
  pancake-mcp:
    image: kukapay/pancakeswap-poolspy-mcp:latest
    restart: unless-stopped
    env_file: .env
    ports:
      - "8080:8080"
      - "8081:8081" # websocket or MCP port
    volumes:
      - ./data:/app/data

Create a .env file (copy from provided .env.example) and set values:

# .env
RPC_URL=https://bsc-dataseed.binance.org/
CHAIN_ID=56
FACTORY_ADDRESS=0xca143ce32fe78f1f7019d7d551a6402fc5350c73  # PancakeSwap factory
WEBHOOK_URL=https://your-server.example/hooks/pool-created
TELEGRAM_BOT_TOKEN=123456:ABC-DEF...
TELEGRAM_CHAT_ID=987654321
DATABASE_URL=sqlite:///data/pools.db
MCP_PORT=8081
APP_PORT=8080
LOG_LEVEL=info

Start the service:

With Docker Compose:

docker-compose up -d

Locally (Node.js):

git clone https://github.com/kukapay/pancakeswap-poolspy-mcp.git
cd pancakeswap-poolspy-mcp
cp .env.example .env   # edit .env
npm install
npm run start

Verify health:

curl http://localhost:8080/health
# { "status": "ok", "uptime": 12.3 }

Available Tools / Resources

The server exposes MCP tools and REST endpoints. Below is a concise reference.

Endpoints (HTTP):

Path	Method	Purpose
/health	GET	Service health
/v1/pools/recent	GET	List recent discovered pools
/v1/pools/{poolAddress}	GET	Get detailed pool info
/v1/alerts/subscribe	POST	Register webhook or subscription
/v1/mcp/tools	GET	List MCP tools available
/v1/mcp/tool/{name}	POST	Invoke MCP tool with arguments

Example MCP tool invocation (tool: get_latest_pools):

curl -X POST "http://localhost:8081/v1/mcp/tool/get_latest_pools" \
  -H "Content-Type: application/json" \
  -d '{"limit":5, "minInitialLiquidityBNB": 0.1}'

Sample response:

[
  {
    "pairAddress": "0xabc123...",
    "token0": { "address":"0x111...", "symbol":"NEW", "decimals":18 },
    "token1": { "address":"0x222...", "symbol":"WBNB", "decimals":18 },
    "initialLiquidityBNB": 0.42,
    "createdAt": "2026-04-08T12:34:56Z",
    "safetyScore": 42
  }
]

Tools exposed to MCP-enabled models:

get_latest_pools (filter by limit, token, min liquidity)
get_pool_details (returns enriched pool and token metadata)
subscribe_alerts (register webhook or return subscription token)
get_stats (historical counts, average initial liquidity)

Use Cases

Alerting and Monitoring: Send webhook or Telegram alerts when pools are created with low initial liquidity or suspicious token metadata, enabling moderators to block potential scams.
Trading Bots: Programmatic feed of new pairs to auto-evaluate and optionally create liquidity or submit initial buy orders based on defined heuristics.
Research & Analytics: Collect time-series data of new listings to analyze trends, token launch characteristics, or market-making behavior.
LLM-Augmented Agents: Provide an MCP tool to a conversational agent so it can fetch the latest pool context during a user query like “Which new tokens just listed with > 1 BNB liquidity?”
Onboarding & Token Discovery: Build a discovery page or dashboard that showcases newly created pairs filtered by volume, verified token status, or developer whitelists.

Notes & Best Practices

Rate limiting and RPC provider reliability matter — use a stable BSC RPC provider for production.
Persisting data in a managed Postgres instance is recommended for scale and analytics; SQLite is fine for local testing.
Configure alert filters (min liquidity, verified contracts, known token lists) to reduce noise and avoid false positives.
Secure webhook endpoints and access tokens; rotate Telegram/Discord tokens if compromised.

For full source, configuration examples and advanced deployment instructions, refer to the project repository on GitHub: https://github.com/kukapay/pancakeswap-poolspy-mcp

PancakeSwap New Liquidity Pools Tracker MCP Server