MCP.so
Sign In
O

Oneqaz Trading Mcp

@oneqaz-trading

Repository

About Oneqaz Trading Mcp

Self-verifying trading MCP — regime detection + Trust Layer for crypto, US, KR stocks. 13 trust tools let AI agents verify accuracy, backtest tuning, news causality, feature governance, and strategy performance before recommending. 1,100+ symbols, 24/7 live data.

#trading#signals#regime-detection#trust-layer#crypto#stocks#fear-greed#fastmcp#b2ai#market-intelligence

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "oneqaz-trading-mcp": {
      "command": "docker",
      "args": [
        "build",
        "-t",
        "oneqaz-trading-mcp",
        "."
      ]
    }
  }
}

Tools

4

market_id, limit, action_filter, min_pnl, max_pnl, hours_back

market_id, min_roi, max_roi, strategy, sort_by, sort_order, limit

market_id, symbol, min_score, max_score, action_filter, interval

market_id, limit, decision_filter, hours_back

Overview

What is Oneqaz Trading Mcp?

Oneqaz Trading Mcp is an MCP server that provides financial market intelligence, including regime detection, self-correcting signals, and macro context across crypto, US stocks, and Korean stocks. It is designed for AI agents and developers who need market-aware context without building the data pipeline themselves.

How to use Oneqaz Trading Mcp?

Connect via a live cloud API at https://api.oneqaz.com/mcp (no install) or run locally using pip install oneqaz-trading-mcp then oneqaz-trading-mcp init and oneqaz-trading-mcp serve. Configure via environment variables such as MCP_SERVER_PORT and DATA_ROOT. After starting, connect any MCP client (e.g., Claude) to the URL http://localhost:8010/mcp.

Key features of Oneqaz Trading Mcp

  • Regime detection (trending, ranging, volatile) per market and globally.
  • Self-correcting signals using Thompson Sampling on trade outcomes.
  • Macro context chain from global regime down to individual symbols.
  • 19 stateless resources and 4 tools for querying signals, positions, decisions.
  • _llm_summary field on every response optimized for AI consumption.
  • 24/7 live cloud API with 1,100+ symbols monitored.

Use cases of Oneqaz Trading Mcp

  • Give AI agents live market awareness with regime and signal context.
  • Build a regime-aware trading system that adapts actions to market state.
  • Perform cross-market macro-to-micro analysis of global flows into assets.
  • Ask Claude directly about market conditions, positions, and risks.

FAQ from Oneqaz Trading Mcp

How is Oneqaz Trading Mcp different from typical financial MCPs?

Oneqaz adds regime detection, self-correcting signals weighted by real outcomes, macro-to-asset context chains, and a live 24/7 cloud API, which typical financial MCPs lack.

What are the dependencies and runtime requirements?

For local use, install via pip; Python and SQLite are used. The live API requires no install beyond an MCP client. Docker is also supported.

Where does the data live?

Data is stored in a configurable root directory (DATA_ROOT) with subdirectories for global regime, coin market, KR market, US market, and external context. The live API uses OneQAZ's cloud servers.

What are the rate limits for the live API?

The live API has a daily quota of 1,500 requests per IP and a burst limit of 30 requests per minute per IP. Exceeding limits returns HTTP 429. Local servers have no rate limits.

What transports and authentication are supported?

The server uses HTTP as its transport (stateless). No authentication is mentioned for the free-tier live API; local demo data requires no auth.

Frequently asked questions

How is Oneqaz Trading Mcp different from typical financial MCPs?

Oneqaz adds regime detection, self-correcting signals weighted by real outcomes, macro-to-asset context chains, and a live 24/7 cloud API, which typical financial MCPs lack.

What are the dependencies and runtime requirements?

For local use, install via pip; Python and SQLite are used. The live API requires no install beyond an MCP client. Docker is also supported.

Where does the data live?

Data is stored in a configurable root directory (`DATA_ROOT`) with subdirectories for global regime, coin market, KR market, US market, and external context. The live API uses OneQAZ's cloud servers.

What are the rate limits for the live API?

The live API has a daily quota of 1,500 requests per IP and a burst limit of 30 requests per minute per IP. Exceeding limits returns HTTP 429. Local servers have no rate limits.

What transports and authentication are supported?

The server uses HTTP as its transport (stateless). No authentication is mentioned for the free-tier live API; local demo data requires no auth.

Comments

Basic information

Category

Finance & Commerce

Transports

stdio

Author

oneqaz-trading

Submitted by

Junhyeong Park

More Finance & Commerce MCP servers

GoLeasy MCP

Der GoLeasy MCP-Server verbindet KI-Assistenten direkt mit der GoLeasy-Vergleichsplattform für Leasing-Angebote. Einfach im Chat beschreiben, was man sucht, anstatt selbst zu suchen und zu filtern - die KI ruft den Serve

U

Unquant

Streetbeat

Hosted financial-data MCP for AI agents: 28 tools for market and company data, financial news, macroeconomics, and US Congress disclosures. Streamable HTTP with an anonymous Free allowance plus OAuth. Delayed/end-of-day

Exact Online

iWebDevelopment

Exact AI Connect is a remote MCP server that connects AI assistants like Claude, ChatGPT, Copilot, Gemini and Perplexity directly to your Exact Online administration. Ask questions about your finances in plain language a

GallanDigital Marketplace

GallanDigital

Read-only discovery server for GallanDigital, a B2B marketplace of credentialed operational systems and tools for PE operating partners, boutique consultants, and business owners. Exposes search, listing detail, contribu

Cello MCP

Cello

Connect Cello MCP to build and scale your referral, partner, or affiliate programs without leaving the conversation. For growth and marketing teams: Ask which referrers drive the most revenue, identify who to re-engage