# TradeOS (test)

> TradeOS (test) is an MCP server for sports data. MCP bridge giving AI clients ticker search, chart technical analysis and macro news (test. It can run locally on your machine or as a hosted remote server.

- **Kind:** MCP server
- **Platforms:** API, CLI, Web, self-hosted
- **Website:** https://ai.tradeos.xyz
- **Repository:** https://github.com/tradeos-ai/tradeos-mcp
- **MCP Registry name:** io.github.TradeOS-AI/tradeos-mcp-test
- **Categories:** [sports data](https://ai-powered.com/categories/sports-data), [trading analysis](https://ai-powered.com/categories/trading-analysis)
- **Listing updated:** 2026-11-18

## About

The README describes TradeOS as a set of agentic technical analysis tools for XAUUSD and other markets, plus no-code AI trading agents that send alerts. This MCP server is the connector that exposes those capabilities to AI clients such as Claude.

The registry entry lists tools for ticker search, a custom agent call, chart technical analysis and macro news. It can be run locally as an npm package over stdio with a bearer token obtained through an OAuth step, or connected remotely over streamable HTTP at the TradeOS endpoint.

The README also covers a Claude Code plugin wrapper with an /tradeos:analyze skill. The registry name and the package are marked as a test build, so the interface may change. Access requires a TradeOS sign-in.

## Features

- **Ticker search:** Looks up market symbols through the search_tickers tool listed in the README.
- **Technical analysis:** Runs chart-based technical analysis via the technical_analysis tool.
- **Macro and news context:** Provides market news and macro context to pair with price analysis.
- **Custom trading agents:** Exposes a customize-agent tool for configuring TradeOS agents that run alerts.
- **Dual transport access:** Available as a local npm stdio package or a remote streamable HTTP endpoint.

## Good for

- Asking an AI assistant for a technical read on gold (XAUUSD) charts
- Pulling macro and news context alongside price analysis
- Setting up alerts through a no-code trading agent from within an AI client
- Comparing spreads or searching for market tickers during research

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http tradeos-mcp-test https://ai.tradeos.xyz/api/agent/mcp-call
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://ai.tradeos.xyz/api/agent/mcp-call
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "tradeos-mcp-test": {
      "url": "https://ai.tradeos.xyz/api/agent/mcp-call"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "tradeos-mcp-test": {
      "type": "http",
      "url": "https://ai.tradeos.xyz/api/agent/mcp-call"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "tradeos-mcp-test": {
      "serverUrl": "https://ai.tradeos.xyz/api/agent/mcp-call"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "tradeos-mcp-test": {
      "type": "http",
      "url": "https://ai.tradeos.xyz/api/agent/mcp-call"
    }
  }
}
```

## Tools

- `mcp_health`: Health check for the MCP connection, named in the README.
- `search_tickers`: Ticker search, named in the README and registry.
- `customize-agent`: Configures a TradeOS agent, named in the README.
- `technical_analysis`: Chart technical analysis, named in the README.
- `bloomberg-oracle-terminal`: Named in the README's tool list; purpose not described.

## FAQ

### How do I authenticate?

The npm package requires a TRADEOS_ACCESS_TOKEN obtained by running the package's oauth command and pasting the printed token. The README says that in Claude, you connect via Settings and sign in.

### Is it open source?

The README links an MIT license file in the repository, but the TradeOS service itself is a hosted product.

### How do I add TradeOS (test) to Claude Code?

Run this command in your terminal: claude mcp add --transport http tradeos-mcp-test https://ai.tradeos.xyz/api/agent/mcp-call

## Where this comes from

- Description: written from the project's README, the project's website and its MCP Registry entry.
- Install steps: from the official MCP Registry entry.

---
Source: https://ai-powered.com/mcp/tradeos-test
