curistat-mcp
Integrates with CrewAI agents as volatility analysis tools, allowing AI analysts to access market forecasts, regime detection, rare signals, and research capabilities for futures trading intelligence.
Provides LangChain agents with volatility analytics tools for futures markets, including forecast generation, signal scanning, directional bias analysis, and historical market research capabilities.
Curistat MCP Server
Futures-specific volatility intelligence for AI agents via the Model Context Protocol.
The first MCP server dedicated to futures volatility analytics. Provides forecasts, regime detection, rare signals, directional bias, and research tools for ES, NQ, and related products.
Quick Start
1. Install
pip install curistat-mcp2. Get a Curistat API key
Sign up at curistat.com/developer and generate an API key.
3. Configure Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"curistat": {
"command": "python",
"args": ["-m", "curistat_mcp"],
"env": {
"CURISTAT_API_KEY": "cst_agent_your_key_here"
}
}
}
}Restart Claude Desktop. You can now ask Claude questions like:
"What's the volatility forecast for ES today?"
"Are there any rare signals firing for NQ?"
"Show me the economic calendar for next week"
"Find historical days similar to today for ES"
Related MCP server: BotSpot
Available Tools
Core Intelligence
Tool | Description |
| Daily volatility forecast (1-10 rating) with expected move ranges |
| 5-day forward volatility forecast |
| Scan all 24 rare volatility signals |
| Directional bias (17-component aggregate) |
| Full session planner with trade setups |
Regime
Tool | Description |
| CRC composite (0-100, 5 bands) |
| Market conditions snapshot |
Research
Tool | Description |
| Economic calendar with vol impact estimates |
| Historical vol analysis around specific events |
| Historical analog day lookup |
Supported Products
ES, NQ, MES, MNQ, SPX, SPY, QQQ
Agent Framework Integration
CrewAI
from crewai import Agent
from crewai_tools import MCPTool
curistat_tools = MCPTool(
server_command="python",
server_args=["-m", "curistat_mcp"],
server_env={"CURISTAT_API_KEY": "cst_agent_..."},
)
analyst = Agent(
role="Futures Volatility Analyst",
tools=[curistat_tools],
goal="Analyze market conditions and provide trading intelligence",
)LangChain
from langchain_mcp import MCPToolkit
toolkit = MCPToolkit(
server_command="python",
server_args=["-m", "curistat_mcp"],
server_env={"CURISTAT_API_KEY": "cst_agent_..."},
)
tools = toolkit.get_tools()Direct stdio (any MCP client)
CURISTAT_API_KEY=cst_agent_... python -m curistat_mcpThe server communicates via stdio using the MCP protocol.
Environment Variables
Variable | Required | Default | Description |
| Yes | -- | Your Curistat Developer API key |
| No |
| API base URL |
Rate Limits
Rate limits are enforced by your API key tier, not by the MCP server. The server is a thin client that proxies requests to the Curistat API.
Tier | Requests/min | Requests/day |
Free | 10 | 100 |
Pro | 60 | 5,000 |
Agent | 120 | 50,000 |
License
Available Tools
10 toolsget_calendarA
Get the economic calendar for the next N days (1-30).
Returns upcoming events with historical volatility impact estimates so agents can anticipate high-impact sessions.
Inputs: days.
get_directionA
Get directional bias from a 17-component signal aggregate.
Products: ES, NQ, MES, MNQ, SPX, SPY, QQQ. Combines breadth, momentum, positioning, and structural signals into a single directional view.
Inputs: product, date.
get_event_impactB
Analyze historical volatility impact of a specific economic event.
event: Event name (e.g. CPI, FOMC, NFP, PPI, Retail Sales). Products: ES, NQ, MES, MNQ, SPX, SPY, QQQ. Returns pre/post event vol stats across multiple historical occurrences.
Inputs: event (required), product.
get_forecast_todayB
Get today's volatility forecast with a 1-10 rating and expected move ranges.
Products: ES, NQ, MES, MNQ, SPX, SPY, QQQ. Returns rating, expected high/low, regime context, and key events.
Inputs: product.
get_forecast_weekA
Get 5-day forward volatility forecast.
Products: ES, NQ, MES, MNQ, SPX, SPY, QQQ. Optional date parameter (YYYY-MM-DD) to anchor the forecast.
Inputs: product, date.
get_pulseB
Get a market conditions snapshot for a product.
Products: ES, NQ, MES, MNQ, SPX, SPY, QQQ. Quick overview of current vol, regime, and key metrics.
Inputs: product, date.
get_regimeA
Get the CRC (Curistat Regime Composite) reading, a 0-100 score across 5 bands that fuses HMM, VPIN, BOCD, Hurst exponent, and Sample Entropy into a single regime indicator.
Inputs: no named parameters.
get_session_planB
Get a full session planner with pattern-driven trade setups for today.
Products: ES, NQ, MES, MNQ, SPX, SPY, QQQ. Returns key levels, expected behavior windows, and actionable setups based on the day's volatility forecast and regime.
Inputs: product, date.
get_signalsA
Scan all 24 rare volatility signals for a given product and date.
Products: ES, NQ, MES, MNQ, SPX, SPY, QQQ. These are statistically rare conditions (VIX extremes, term structure inversions, volatility compression, etc.) that precede outsized moves.
Inputs: product, target_date.
get_similar_daysA
Find historical days with conditions similar to today.
Products: ES, NQ, MES, MNQ, SPX, SPY, QQQ. Uses feature-vector similarity to find analog days, then shows what happened next -- useful for probabilistic scenario planning.
Inputs: product, limit.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 10 tools
Each tool has a distinct purpose: calendar, directional bias, event impact, daily/weekly forecasts, market pulse, regime, session plan, signals, and similar days. No two tools overlap in functionality.
All tools follow a consistent 'get_' prefix with clear noun descriptors (e.g., get_calendar, get_direction, get_event_impact). No mixed conventions or vague names.
10 tools is well-scoped for a market analysis server. Each tool provides a specific, valuable capability without excess or redundancy.
The tool set covers the full spectrum of market analysis: forecasts, regime, events, signals, and historical analogs. No obvious gaps within the stated domain of volatility and directional analysis.
Maintenance
Related MCP Connectors
Production MCP server for US equity and options intelligence: real-time IV radar, Monte Carlo simulation, options pressure, strategy backtesting, AI prediction, pre-trade risk analysis, and automated stock research reports.
Research-only MCP server: your AI as a quant research desk. 90 tools, no trades, no brokers.
Trading MCP server for AI agents, with live market data, account reads and controlled execution.
Official Intrinio MCP server. US stock prices, options and fundamentals for AI agents.
Related MCP Servers
- -licenseNot gradedqualityNot gradedmaintenanceReal-time financial market data MCP server. Stocks, crypto, technicals, sentiment, FDA calendar. No API keys required.-
- FlicenseNot gradedqualityBmaintenanceFull-lifecycle algorithmic trading MCP server. AI strategy generation from plain English, backtesting, live bot deployment to 10+ brokers, portfolio monitoring, and prediction markets. Stocks, options, crypto, futures. 32 tools. Free tier.-
- AlicenseAqualityDmaintenanceAn MCP server that exposes futures prop firm data and personalized recommendation logic to AI agents like Claude and ChatGPT.6MIT
- FlicenseNot gradedqualityDmaintenanceMCP server for domestic futures analytics, providing tools to fetch OHLCV data, check data quality, analyze market metrics, and generate markdown reports via the Model Context Protocol.-