Skip to main content
Glama
Markolis MCP Server

Markolis MCP Server

Official

Markolis MCP Server

Institutional Quantitative Market Intelligence & Micro-Apps for AI Agents

Official MCP Registry License: Apache-2.0 Python 3.9+ Node 18+ Status

Terminal & Web App • API Documentation • Official Registry


Overview

Markolis brings institutional quantitative telemetry, 100-bin Auction Market Theory (AMT), order book trapped crowd sentiment, and distribution-free conformal volatility cones to autonomous AI agents and trading desks.

Instead of relying on lagging retail indicators (RSI, MACD, Moving Averages), Markolis queries empirical order book volume distributions across 130+ global financial assets (Gold, Bitcoin, S&P 500, Nasdaq, EUR/USD, Crude Oil, and FX crosses).

Key Capabilities

  • 🎯 100-Bin Auction Market Theory: Point of Control (POC), 70% Value Area (VAH / VAL), and balance rotation regimes.

  • 🪤 Order Book Trapped Crowd Radar: Net long/short retail cluster offside percentages and nearest stop-sweep vectors.

  • ⚡ Conformal Volatility Cones: Distribution-free 90% confidence excursion bands calibrated across market sessions.

  • 🌐 Global Macro Radar: Real-time cross-asset session overlaps (London / NY / Tokyo), liquidity multipliers, and news blackouts.

  • 🖥️ Interactive MCP Micro-Apps (ui://markolis/market-frame): Renders interactive HTML5 visual auction frames inside AI chat clients.


Related MCP server: north7-market-intelligence

Quickstart

You can connect Markolis to your favorite AI assistant in seconds:

No software installation or local dependencies required. Directly configure your MCP client to stream from the hosted endpoint:

1. Claude Desktop

Add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "markolis": {
      "url": "https://www.markolis.com/api/mcp"
    }
  }
}

Config file locations:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

  • Windows: %APPDATA%\Claude\claude_desktop_config.json

2. Cursor

Open Cursor Settings $\rightarrow$ Features $\rightarrow$ MCP $\rightarrow$ + Add New MCP Server:

  • Name: markolis

  • Type: SSE (or HTTP)

  • Server URL: https://www.markolis.com/api/mcp

Or in your project's .cursor/mcp.json:

{
  "mcpServers": {
    "markolis": {
      "url": "https://www.markolis.com/api/mcp"
    }
  }
}

3. Windsurf & Zed

In your MCP configuration:

{
  "mcpServers": {
    "markolis": {
      "url": "https://www.markolis.com/api/mcp"
    }
  }
}

Option B: Node.js STDIO Runner (npx)

If your MCP client requires a local stdio command:

{
  "mcpServers": {
    "markolis": {
      "command": "npx",
      "args": ["-y", "markolis-mcp"]
    }
  }
}

Option C: Python STDIO Runner (uvx / python)

{
  "mcpServers": {
    "markolis": {
      "command": "uvx",
      "args": ["markolis-mcp"]
    }
  }
}

Or with standard Python:

pip install markolis-mcp
python -m markolis_mcp

Available MCP Tools

Tool

Parameters

Description

render_market_intelligence_frame

query (str, required)default_symbol (str, default: 'XAU_USD')

Queries market intelligence and returns both empirical quantitative analysis and an interactive visual micro-app frame (ui://markolis/market-frame).

get_market_telemetry

symbol (str, required)

Retrieves comprehensive quantitative metrics: POC, Value Area, Hurst persistence exponent, order book crowd imbalance, and Garman-Klass volatility.

get_macro_radar

None

Retrieves real-time global macroeconomic radar feeds: active session clocks, liquidity multipliers, upcoming news blackouts, and offshore USD velocity.

get_prop_guardian_status

None

Live systematic risk telemetry for prop challenge pacing: trailing drawdown headroom, daily loss usage, and Bayesian Kelly allocation sizing.

get_verified_catalog

None

Full 26-stream verified asset catalog with Garman-Klass volatility ratings and asset classes.


Example Prompts

Once installed in Claude or Cursor, simply ask:

  • "Where is the retail crowd trapped on Gold ($XAUUSD)?"

  • "Show me the 100-bin auction theory profile and Value Area for Bitcoin ($BTCUSD)."

  • "What is the active macroeconomic session overlap and liquidity multiplier right now?"

  • "Calculate the 90% conformal volatility cone for EUR/USD heading into the New York session."


Interactive MCP Micro-Apps

Markolis implements the Model Context Protocol UI Specification (ui://markolis/market-frame). When calling render_market_intelligence_frame, supporting clients render an interactive HTML5 frame containing:

  1. Interactive 100-Bin Auction Histogram: Real-time volume profiles highlighting Value Area High (VAH), Value Area Low (VAL), and Point of Control (POC).

  2. Order Book Trapped Crowd Sentiment: Visual long vs. short positioning and stop-sweep risk vectors.

  3. Session Regime Sparklines: Regime persistence and volatility cone bands.


Quotas & Commercial Tiers

Markolis provides a generous free tier (25 requests/day) with zero credit card or registration required.

For institutional desks, algorithmic trading agents, and quantitative researchers:

  • Free: 25 queries/day, 131 assets, full MCP access.

  • Pro ($49/mo): 300 queries/day, private agent tokens, continuous webhook feeds, and priority execution.

  • Enterprise / Prop Desks: Dedicated custom streams, Sub-10ms WebSockets, and custom data lakes.

Visit https://www.markolis.com to upgrade or manage keys.


Regulatory Compliance & Disclaimer

CFTC Rule 4.41 Compliance: All quantitative metrics, excursions, order book profiles, and volatility cones delivered by Markolis are empirical statistical estimates based on market structure snapshots. Markolis operates strictly as a quantitative intelligence engine and does not provide financial advice, trading recommendations, or buy/sell calls.


License

This project is licensed under the Apache License 2.0.

Related MCP Connectors

Related MCP Servers

  • F
    license
    Not graded
    quality
    C
    maintenance
    Enables MCP-compatible AI clients to access live crypto market data and AI-driven quantitative analysis, with structured outputs and full observability.
    -
  • A
    license
    A
    quality
    C
    maintenance
    AI-powered financial intelligence for autonomous trading agents. 9 MCP tools for real-time trading signals, risk index, market regime detection, stock analysis, commodity scoring, sector radar, and geopolitical intelligence briefings.
    13
    1
    MIT
  • A
    license
    Not graded
    quality
    C
    maintenance
    Enables MCP-compatible AI agents to retrieve quantum-computed portfolio optimisation, VaR simulations, AI-enhanced sentiment and regime detection, and cross-asset risk signals through simple metric tools.
    49 npm
    MIT
  • F
    license
    Not graded
    quality
    B
    maintenance
    Financial intelligence MCP server exposing real market data, SEC filings, FinBERT sentiment, technical analysis, and macro/global market data as MCP tools — free tier issued instantly, no card required.
    -