mcp-server-sentinel

English | 繁體中文 | 简体中文 | 日本語 | 한국어

MCP server for Sentinel Bot — algorithmic trading backtesting, bot management, and account operations via AI agents.

This server implements the Model Context Protocol (MCP), providing 36 tools that let AI agents run crypto backtests, deploy trading bots, optimize parameters, browse strategy marketplace, and manage accounts & payments — all through natural language.

Why Use This MCP Server?

This MCP server is built for Sentinel — an autonomous algorithmic trading platform designed for AI agent workflows. By connecting your AI agent to Sentinel via MCP, you get a complete agent-driven trading pipeline:

• Natural language backtesting — describe a strategy idea, AI generates and tests it
• One-conversation deployment — go from backtest to live trading bot without writing code
• Real-time monitoring — PnL tracking, trading signals, and SMC market analysis
• Full autonomy — strategy research, parameter optimization, and trade execution, all handled by your AI agent

Best paired with Claude Code, Claude Desktop, or any MCP-compatible AI client.

Quick Start

1. Get an API Key

Sign up for free at sentinel.redclawey.com. All plans (including the free trial) support API Keys. Generate a key in your dashboard under Settings > API Keys. Purchase credits to start backtesting, then upgrade your plan as needed for more bots.

2. Installation

Claude Code (Recommended)

claude mcp add sentinel -- npx mcp-server-sentinel

Then set your environment variable:

export SENTINEL_API_KEY=sk-your-api-key-here

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "sentinel": {
      "command": "npx",
      "args": ["-y", "mcp-server-sentinel"],
      "env": {
        "SENTINEL_API_KEY": "sk-your-api-key-here"
      }
    }
  }
}

Manual Setup (Any MCP Client)

{
  "command": "npx",
  "args": ["-y", "mcp-server-sentinel"],
  "env": {
    "SENTINEL_API_KEY": "sk-your-api-key-here",
    "SENTINEL_API_URL": "https://sentinel.redclawey.com/api/v1"
  }
}

Environment Variables

Tools (36)

Backtesting

Bot Management

Exchanges

OKX Exchange

AI Strategy

Grid Optimization

Analysis & Signals

Strategy Marketplace

Account & Payments

Usage Examples

Typical AI agent workflow:

1. build_strategy          -> AI generates strategy from description (1 credit)
2. run_backtest            -> Validate on historical data
3. run_grid_backtest       -> Optimize parameters
4. get_grid_results        -> Find best parameter combo
5. create_bot              -> Deploy with optimal params
6. start_bot               -> Go live
7. get_signals             -> Monitor signal execution
8. get_analysis            -> Check daily market analysis
9. get_bot_performance     -> Review PnL

When credits run out:

get_account_info           -> credits_balance: 0
create_payment_link        -> strategy_id: "credits_topup", amount_usd: 20
  -> Returns payment URL -> User completes payment in browser
verify_payment             -> Confirm credits added
run_backtest               -> Continue backtesting

Supported Symbols

BTC, ETH, SOL, XRP, BNB, DOGE, LINK, TRX, SUI

Supported Timeframes

1m, 5m, 15m, 1h, 4h, 1d

Entry Block Types

Exit Block Types

Plan Pricing

All plans include API Key access. Start with the free trial, purchase credits for backtesting, then upgrade your plan as needed for more bots or symbols.

Credits are consumed per backtest run. Use create_payment_link or create_crypto_invoice with strategy_id: "credits_topup" to top up (minimum $10, 17 credits per $1).

Payment Methods

• Credit Card / USDC: Via Helio (create_payment_link) — Visa, Mastercard, or USDC on Solana
• 300+ Cryptocurrencies: Via NOWPayments (create_crypto_invoice) — BTC, ETH, SOL, USDT, etc.

Development

git clone https://github.com/clarencyu-boop/mcp-server-sentinel.git
cd mcp-server-sentinel
npm install
SENTINEL_API_KEY=sk-your-key node src/index.mjs

License

MIT