Skip to main content
Glama

VibeTrader MCP Server


A Model Context Protocol (MCP) server that connects AI assistants like Claude Desktop, Cursor, and Windsurf to VibeTrader's trading platform.

Features

  • Create Bots: Create AI-powered trading bots using natural language

  • Manage Bots: Start, pause, delete, and monitor your bots

  • Portfolio: View positions and account balance

  • Market Data: Get quotes and options chains

  • Backtesting: Test strategies on historical data

Related MCP server: Alpaca MCP Server

Available Tools

All 30 are available over either endpoint below.

Tool

Description

authenticate

Authenticate with your API key from vibetrader.markets/settings

create_bot

Create a new trading bot using natural language

list_bots

List all your trading bots

search_bots

Search for bots by name

get_bot

Get details of a specific bot

update_bot

Update a bot's name or status

refine_bot

Refine a bot's strategy using natural language

start_bot

Start a paused bot

pause_bot

Pause a running bot

delete_bot

Delete a bot

start_all_bots

Start all paused bots at once

pause_all_bots

Pause all running bots at once

delete_all_bots

Delete ALL bots (requires confirm='DELETE ALL')

get_bot_strategy

Get the full strategy configuration for a bot

explain_bot_strategy

Get a plain English explanation of a bot's strategy

get_bot_stats

Get aggregate statistics across your bots

get_daily_trade_stats

Get daily trading statistics for a bot

get_portfolio

Get portfolio positions and balance

get_positions

Get current open positions with unrealized P&L

get_account_summary

Get account balance, buying power, and key metrics

close_position

Close a position by symbol

place_order

Place a buy or sell order

get_trade_history

Get recent trade history across all bots

get_bot_trades

Get trade history for a specific bot

get_quote

Get current quote for a stock

get_market_status

Check if US stock markets are currently open

run_backtest

Run a historical backtest for a bot's current strategy

list_backtests

List all backtests for a bot

get_backtest

Get detailed results for a specific backtest

quick_covered_call_backtest

Run a quick covered call income strategy backtest

Endpoints

The hosted server speaks both MCP HTTP transports:

URL

Speaks

https://vibetrader-mcp-289016366682.us-central1.run.app/mcp

Streamable HTTP — the current MCP transport. Use this one. It is also what the in-app setup at vibetrader.markets/settings gives you.

https://vibetrader-mcp-289016366682.us-central1.run.app/sse

Either transport. A client that opens an SSE stream gets one (with its companion /messages/ POST route, which the stream advertises — you never configure it by hand); a client that POSTs JSON-RPC gets streamable HTTP, the same session space as /mcp.

Both URLs expose the same 30 tools and the same authentication, so an existing /sse setup keeps working as-is. Prefer /mcp for anything new: SSE is deprecated upstream, and /sse is kept working for setups copied from older versions of this README.

Setup for Claude Desktop

  1. Get your API token from https://vibetrader.markets/settings

  2. Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "vibetrader": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://vibetrader-mcp-289016366682.us-central1.run.app/mcp"]
    }
  }
}

Clients that connect to a remote MCP URL directly (Cursor, Windsurf, and Claude Desktop's own remote-server support) can use the URL on its own:

{
  "mcpServers": {
    "vibetrader": {
      "url": "https://vibetrader-mcp-289016366682.us-central1.run.app/mcp"
    }
  }
}

/sse works too — see Endpoints — but /mcp is the one to use for a new setup.

  1. Send your token as a header. Against the hosted server this is how you authenticate — see the note below:

{
  "mcpServers": {
    "vibetrader": {
      "url": "https://vibetrader-mcp-289016366682.us-central1.run.app/mcp",
      "headers": { "Authorization": "Bearer vt_YOUR_TOKEN" }
    }
  }
}
  1. Restart Claude Desktop

  2. Create bots:

    "Create a trading bot that buys AAPL when RSI goes below 30"

A note on the authenticate tool

There is an authenticate tool, and over a local stdio setup (python server.py stdio) it does what you would expect: hand it your token once and the rest of the session is authenticated.

Over HTTP — which is every hosted setup above — use the Authorization header instead. The header is resolved per request and stored nowhere, so it is also the safer of the two: the server never holds your token at all.

Example Conversations

Create a momentum bot:

"Create a trading bot that buys AAPL when RSI crosses below 30 and sells when it crosses above 70"

Check your bots:

"Show me all my trading bots and their performance"

Backtest a strategy:

"Backtest a moving average crossover strategy on SPY for the last 6 months"

Options trading:

"Create an iron condor bot for SPY that enters when IV rank is above 50"

Development

# Install dependencies
pip install -r requirements.txt

# Run locally (HTTP mode — serves both /mcp and /sse on :8080,
# the same app the deployment runs)
python server.py

# ...or exactly as the container does
uvicorn server:app --host 0.0.0.0 --port 8080

# Run in STDIO mode (for a local Claude Desktop install)
python server.py stdio

Then point a client at http://localhost:8080/mcp or http://localhost:8080/sse.

Deployment

Deployed on Google Cloud Run at mcp.vibetrader.markets

gcloud run deploy vibetrader-mcp --source . --region us-central1

Security

  • Your API token is required for authentication

  • All requests are made over HTTPS

  • Tokens are resolved per request and never shared between users: over HTTP the Authorization: Bearer vt_... header is read from your own request and stored nowhere at all

  • Paper trading mode is default for safety

Related MCP Connectors

Related MCP Servers

  • A
    license
    Not graded
    quality
    D
    maintenance
    Enables AI agents to execute stock trading operations with built-in risk controls and human approval workflows. Supports paper trading simulation, real brokerage integration (Alpaca, Tradier), backtesting, sentiment analysis, and portfolio management while maintaining strict separation between AI intelligence and trade execution.
    MIT
  • A
    license
    Not graded
    quality
    C
    maintenance
    Enables natural language trading operations through AI assistants using Alpaca's Trading API. Supports stocks, options, crypto trading, portfolio management, and real-time market data access.
    MIT
  • A
    license
    Not graded
    quality
    D
    maintenance
    Enables natural language trading operations for stocks, options, crypto, and portfolio management via Alpaca's Trading API through AI assistants.
    MIT
  • A
    license
    A
    quality
    A
    maintenance
    Provides AI assistants with market screening, analysis, and scoring across stocks, crypto, and forex, enabling natural language queries for trading insights.
    4
    MIT