Market Intelligence MCP
Official市场情报 MCP 服务器
衰退概率、资本轮动、宏观连锁反应情景分析、投资计算器以及实时经济数据 — 适用于 Claude、ChatGPT、Cursor 及任何 MCP 客户端。
由 BullrunData API 提供支持。
快速入门
Claude Desktop
添加到你的 claude_desktop_config.json:
{
"mcpServers": {
"market-intelligence": {
"command": "npx",
"args": ["-y", "@bullrundata/market-intelligence"],
"env": {
"BULLRUNDATA_API_KEY": "your-api-key"
}
}
}
}Claude Code
claude mcp add market-intelligence -- npx -y @bullrundata/market-intelligence获取你的 API 密钥
在 bullrundata.com 免费注册 — 每天 100 次调用,无需信用卡。
Related MCP server: FeedOracle Macro MCP
可用工具 (23)
衰退情报
工具 | 描述 |
| 基于 15 个指标加权模型的当前衰退概率 (0-95%) |
| 包含分数和数值的所有模型指标 |
| 美联储政策立场及权重影响 |
| 市场周期阶段(早期/中期/晚期/衰退) |
| 确认或否定信号的 4 个同步指标 |
| 萨姆规则 (Sahm Rule) 计算 — 预测了 1970 年以来的每一次衰退 |
资本轮动
工具 | 描述 |
| 基于 9 种宏观工具的风险偏好/厌恶综合得分 (-100 到 +100) |
| 包含价格、信号和分数的所有工具 |
| 相关性破裂和动量转换 |
投资计算器
工具 | 描述 |
| 租赁房产分析:资本化率、偿债覆盖率 (DSCR)、现金流、1% 规则 |
| BRRRR 交易评分 (0-100),包含 70% 规则和完整细分 |
经济数据
工具 | 描述 |
| 按系列 ID 查询任何经济指标(GDP、CPI、UNRATE 等) |
| 按关键词搜索可用指标 |
| 联邦基金利率、国债收益率、抵押贷款利率 |
| CPI、PCE、盈亏平衡通胀率、通胀预期 |
| 失业率、非农就业、申请人数、JOLTS |
| 抵押贷款利率、开工率、许可数、房价 |
| 收益率曲线利差和倒挂检测 |
| VIX、金融状况、压力指数、消费者信心 |
连锁反应引擎(宏观情景分析)
工具 | 描述 |
| 列出所有 10 种宏观催化剂情景 |
| 带有实时数据增强的催化剂完整连锁反应分析 |
| 按关键词搜索催化剂(例如:"oil", "dollar", "china") |
| 按类别筛选:地缘政治、货币、信贷、大宗商品、汇率、结构性 |
可用催化剂: 石油供应冲击、美元流动性紧缩、美联储紧急降息、美国衰退、中国台湾局势升级、收益率曲线倒挂、信贷市场冻结、新兴市场货币危机、贸易战升级、主权债务危机
示例对话
“当前的衰退风险是多少?”
Claude 使用 recession_probability 并返回:
当前衰退概率为 37.9%(中度风险)。美联储立场:中性。市场周期:中期。4 个同步指标中 0 个确认。建议:信号混合,需保持谨慎。开始防御性轮动。
“现在是购买 45 万美元投资房产的好时机吗?”
Claude 使用 investment_property_analysis + interest_rates:
以 45 万美元价格、20% 首付、7% 利率计算,月现金流为 287 美元。资本化率:5.3%。现金回报率:4.2%。偿债覆盖率 (DSCR):1.18(健康)。当前 30 年期抵押贷款利率:6.38%。
“目前的市场风险偏好/厌恶信号是什么?”
Claude 使用 capital_rotation_score + divergence_alerts:
资本轮动得分:-3(中性/信号混合)。VIX 升至 27。黄金出现避险买盘。铜价持稳 — 并非结构性破裂。策略:注重质量和多元化,而非集中押注。
“如果油价达到 120 美元会发生什么?”
Claude 使用 cascade_analysis + capital_rotation_score:
石油供应冲击连锁反应:美元需求激增(机械性)-> 亚洲外汇储备流失(可能,数周内)-> 亚洲股市资金外流(可能)-> 亚洲被迫加息(大概率,数月内)。当前资本轮动得分:-13(中性/混合)。策略:通胀对冲(XLE, TIP)和现金储备。关注美联储互换额度使用情况以获取系统性信号。
配置
环境变量 | 必需 | 描述 |
| 是 | 你在 bullrundata.com 获取的 API 密钥 |
定价
层级 | 每日调用次数 | 价格 |
免费版 | 100 | $0 |
专业版 | 10,000 | $29/月 |
商业版 | 100,000 | $99/月 |
支持
许可证
MIT
Available Tools
23 toolsbrrrr_analysisARead-only
Analyze a BRRRR (Buy, Rehab, Rent, Refinance, Repeat) real estate deal. Returns all-in cost, ARV margin, refinance cash-out, monthly cash flow, BRRRR Score (0-100), 70% rule check, and full breakdown.
Inputs: purchasePrice (required), arv (required), monthlyRent (required), rehabCosts, +16 more.
capital_rotation_instrumentsARead-only
Get detailed scoring for all 9 macro instruments in the capital rotation model. Each instrument includes price, weighted score, signal direction, interpretation, and risk classification.
Inputs: no named parameters.
capital_rotation_scoreARead-only
Get the current capital rotation risk-on/risk-off composite score (-100 to +100) based on 9 macro instruments (DXY, Gold, Oil, Copper, VIX, 10Y Yield, Gold/Silver Ratio, SOFR, Bitcoin). Includes regime detection and asset allocation playbook.
Inputs: no named parameters.
cascade_analysisARead-only
Get the full chain reaction cascade for a macro catalyst scenario. Maps cause-effect chains across markets, regions, and asset classes with confidence levels, historical precedents, and live market data enrichment. Use cascade_list to see available catalysts.
Inputs: catalyst_id (required), include_live_data.
cascade_by_categoryARead-only
Get all catalyst scenarios in a specific category. Categories: geopolitical, monetary, credit, commodity, currency, contagion, structural.
Inputs: category (required).
cascade_listARead-only
List all available macro catalyst scenarios in the Cascade Engine. Returns catalyst IDs, names, categories, and severity levels. Use a catalyst ID with cascade_analysis for the full chain reaction.
Inputs: no named parameters.
cascade_searchARead-only
Search catalyst scenarios by keyword. Matches against catalyst names, descriptions, categories, and trigger conditions. Use this when you're not sure which catalyst applies to a given situation.
Inputs: query (required).
confirmation_statusARead-only
Check 4 coincident economic indicators that confirm or deny recession signals: Real GDP Growth, Industrial Production, Real Personal Income, Employment Level.
Inputs: no named parameters.
divergence_alertsARead-only
Get active correlation breakdowns between macro instruments (e.g., DXY and Gold both rising = systemic fear). Includes severity, implication, and momentum shift detection.
Inputs: no named parameters.
economic_indicatorBRead-only
Get time series data for any economic indicator by series ID (e.g., GDP, UNRATE, CPIAUCSL). Returns date-value pairs.
Inputs: series_id (required), limit, start_date, units.
employment_dataARead-only
Get current employment indicators: unemployment rate, nonfarm payrolls, initial claims, JOLTS openings and quits rate, participation rate, average hourly earnings.
Inputs: no named parameters.
fed_stanceARead-only
Get the current Federal Reserve monetary policy stance (TIGHTENING, EASING, CRISIS, or NEUTRAL) based on fed funds rate trajectory and 6-month rate change.
Inputs: no named parameters.
housing_dataARead-only
Get current housing market data: 30Y and 15Y mortgage rates, housing starts, building permits, median sales price, Case-Shiller index, existing home sales.
Inputs: no named parameters.
inflation_dataARead-only
Get current inflation indicators: CPI, Core CPI, PCE, Core PCE, 10-year breakeven inflation, and consumer inflation expectations.
Inputs: no named parameters.
interest_ratesARead-only
Get current interest rates: Fed funds, 2Y/5Y/10Y/30Y Treasury yields, 3M T-Bill, prime rate, 30-year and 15-year mortgage rates.
Inputs: no named parameters.
investment_property_analysisARead-only
Analyze a rental investment property. Returns cap rate, cash-on-cash ROI, monthly cash flow, NOI, DSCR, 1% rule evaluation, and full expense breakdown.
Inputs: purchasePrice (required), monthlyRent (required), downPaymentPct, interestRate, +8 more.
market_regimeARead-only
Get the current market cycle phase: EARLY_CYCLE, MID_CYCLE, LATE_CYCLE, or RECESSION. Based on unemployment trend, ISM PMI, yield curve, and credit spreads.
Inputs: no named parameters.
market_sentimentARead-only
Get market sentiment indicators: VIX (fear index), Financial Conditions Index, Financial Stress Index, and Consumer Sentiment.
Inputs: no named parameters.
recession_indicatorsARead-only
Get all recession model indicators with current values and risk scores.
Inputs: no named parameters.
recession_probabilityARead-only
Get the current US recession probability (0-95%) from a 15-indicator weighted model with dynamic Fed stance and market regime adjustments. Includes risk level, confidence, confirmation status, and actionable recommendation.
Inputs: no named parameters.
sahm_ruleARead-only
Check the Sahm Rule recession indicator. The 3-month moving average of unemployment minus the 12-month minimum. Triggers at 0.50 — has predicted every US recession since 1970.
Inputs: no named parameters.
search_indicatorsARead-only
Search for economic indicators by keyword. Returns matching series with ID, name, frequency, and last updated date.
Inputs: query (required), limit.
yield_curveARead-only
Get yield curve analysis: 10Y-2Y and 10Y-3M spreads, all Treasury yields, and inversion detection.
Inputs: no named parameters.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 23 tools
Each tool targets a distinct aspect of market intelligence—real estate, macro indicators, recession models, yield curves, etc. Potential overlaps like recession_indicators and recession_probability are differentiated by their outputs, with one listing raw indicators and the other providing a composite probability. The cascade suite is clearly organized with separate tools for listing, searching, and analyzing by category.
Most tools follow a noun_noun pattern (e.g., housing_data, recession_probability), but there are deviations: search_indicators uses a verb_noun pattern, and cascade_by_category includes a preposition. This inconsistency is minor and does not impede readability, but a uniform verb_noun or noun_noun convention would improve predictability.
With 23 tools, the server covers a broad domain without being overwhelming. Each tool serves a clearly defined function, and the count aligns well with the scope of market intelligence. Slightly towards the upper end of reasonable, but justified by the diversity of macroeconomic and real estate analysis subdomains.
The toolset thoroughly covers macroeconomic indicators, recession prediction, yield curves, and real estate investment analysis. However, it lacks certain areas like sector-level stock analysis, commodity prices beyond the capital rotation model, or detailed forex data. Overall, core workflows are well-supported, with only minor gaps.
Maintenance
Related MCP Connectors
Live macro data for AI agents: regime, conditions, liquidity, with input provenance. No signup.
macro-snapshot MCP — the state of the economy in one call.
Macro forecasts for inflation, rates, labor, GDP, housing, commodities, currencies, and equities.
Macroeconomic and market data for AI agents: rates, credit, volatility, calendars and screeners.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceUS Macro Economic Intelligence MCP — 8 tools for Fed rates, inflation, yield curve, labor market, GDP via FRED. Part of ToolOracle.-
- FlicenseNot gradedqualityDmaintenanceMacro economic intelligence MCP — 13 tools powered by 86 FRED series + ECB data covering Fed rates, inflation (CPI/PCE), yield curve, GDP, recession probability, and regime classification. Every response ES256K signed.1-
- AlicenseNot gradedqualityDmaintenanceEconomic data MCP server that connects FRED, BLS, BEA, IMF, World Bank, and ECB to any MCP-compatible client, with built-in methodology rules to guide LLMs in selecting appropriate economic indicators.MIT
- AlicenseNot gradedqualityCmaintenanceAgent-ready economic, market & geo-health intelligence — 163 MCP tools, 155 driver-backed indices.MIT