fffinstill
MCP server for fffinstill — A Research Lab for Your Investment Ideas
README
@fffinstill/mcp-server
MCP server for fffinstill — investment analysis for 2,260+ US equities.
Connect Claude Desktop, Cursor, VS Code, or any MCP-compatible AI assistant to fffinstill's stock analysis, screening, macro data, ML earnings predictions, and alternative data tools.
Features
- 69 tools — Stock analysis, screening, portfolio analytics, macro indicators, ML predictions, insider/congressional trades, alternative data, risk metrics, 18 academic factor models, and 13 Business-exclusive tools
- 7 resources — Market signal, macro regime, Fear & Greed, sector overview, company summaries
- 15 prompts — Pre-built workflows for stock analysis, comparison, screening, macro outlook, portfolio health, alternative data, market dashboard, deep value, risk assessment, growth momentum, income analysis, factor models, insider intelligence
Quick Start
1. Get an API Key
Sign up at fffinstill.com and generate an API key at Settings → API.
Requires a Founding, Pro, or Business subscription.
2. Configure Your AI Client
Claude Desktop
Add to ~/.config/claude/claude_desktop_config.json (Linux/macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"fffinstill": {
"command": "npx",
"args": ["-y", "@fffinstill/mcp-server"],
"env": {
"FINSTILL_API_KEY": "finstill_xxx_yyy"
}
}
}
}
Cursor
Add to .cursor/mcp.json in your project root:
{
"mcpServers": {
"fffinstill": {
"command": "npx",
"args": ["-y", "@fffinstill/mcp-server"],
"env": {
"FINSTILL_API_KEY": "finstill_xxx_yyy"
}
}
}
}
VS Code
Add to .vscode/mcp.json in your project root:
{
"servers": {
"fffinstill": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@fffinstill/mcp-server"],
"env": {
"FINSTILL_API_KEY": "finstill_xxx_yyy"
}
}
}
}
3. Use It
Once configured, your AI assistant can:
- "Analyze AAPL" → calls
getCompanyMetrics,getReverseDCF,getBeneishScore, etc. - "Screen for high-quality tech stocks" → uses the
screen-stocksprompt - "Compare NVDA, AMD, and INTC" → calls
compareTickers - "What's the macro outlook?" → uses the
macro-outlookprompt - "Is MSFT likely to beat earnings?" → uses
earnings-previewprompt
CLI Usage
# With npx (no install)
npx @fffinstill/mcp-server --api-key finstill_xxx_yyy
# With environment variable
FINSTILL_API_KEY=finstill_xxx_yyy npx @fffinstill/mcp-server
# Global install
npm install -g @fffinstill/mcp-server
fffinstill-mcp --api-key finstill_xxx_yyy
# Show help
fffinstill-mcp --help
Tools Reference
Stock Analysis (16 tools)
| Tool | Description |
|---|---|
getCompanyMetrics(ticker) |
Full fundamentals: revenue, margins, valuations, quality scores |
getAnalystRatings(ticker) |
Consensus, price targets, ratings breakdown |
getPriceHistory(ticker, weeks?) |
Weekly prices with returns (default: 52 weeks) |
getQuarterlyFinancials(ticker) |
8 quarters: revenue, margins, ROIC, D/E |
getScoreHistory(ticker) |
Composite quality score (7 components) over 12 periods |
getReverseDCF(ticker) |
Bear/base/bull fair value scenarios |
getBeneishScore(ticker) |
M-Score earnings manipulation detection |
getZScoreHistory(ticker) |
Altman Z-Score bankruptcy risk history |
getConvictionHistory(ticker) |
Insider conviction trajectory scores (buying patterns) |
getMacroSensitivity(ticker) |
Sensitivity to rates, inflation, VIX, USD |
getEarningsPrediction(ticker) |
ML earnings beat/miss prediction |
getEstimateRevisions(ticker) |
Analyst EPS revision trends |
getCorrelatedStocks(ticker) |
Price correlation analysis with sector peers |
getInstitutionalOwnership(ticker) |
Top holders, whale activity, concentration risk |
getGeographicRevenue(ticker) |
Revenue by geographic region |
getSegmentRevenue(ticker) |
Business segment revenue breakdown |
Screening & Portfolio (7 tools)
| Tool | Description |
|---|---|
screenStocks(...) |
Multi-factor screen of 2,260+ stocks |
deepValueScreen(sector?, limit?) |
Piotroski ≥ 6 value stocks |
compareTickers(tickers) |
2-5 stocks side-by-side |
computeMetric(metric, aggregation) |
Aggregate statistics across top 100 stocks |
getPortfolioAnalysis(tickers) |
Multi-holding portfolio analytics |
crossReferencePortfolio(tickers) |
Portfolio vs market signal alignment |
getStrategyPicks(strategy?) |
Pre-computed strategy picks |
Market & Macro (13 tools)
| Tool | Description |
|---|---|
getMarketSignal() |
Current signal with 5-component breakdown |
getRegimeAnalysis() |
Macro regime classification |
getMacroIndicator(indicator) |
FRED data (CPI, GDP, rates, etc.) |
getFearGreedIndex(timeline?) |
Fear & Greed with optional history |
getSectorBreakdown(sector?) |
Top stocks by market cap for a sector |
getBLSIndicator(series_id) |
BLS economic data (employment, CPI, PPI) |
getMacroAttention(weeks?) |
Macro Google Search Volume for stress terms |
getRegimeTransitions(days?) |
Historical regime transitions |
getSignalHistory(days?) |
Daily signal history + SPY returns |
getMarketBreadth(period_type?) |
Advance/decline, new highs/lows |
getEarningsCalendar(from?, to?) |
Upcoming earnings reports calendar |
searchStocks(query) |
Fuzzy search 2,260+ US equities |
getMarketHeatmap() |
Sector performance heatmap with top gainers/losers |
Alternative Data (9 tools)
| Tool | Description |
|---|---|
getInsiderActivity(ticker) |
SEC Form 4 insider trades |
getCongressionalTrades(...) |
STOCK Act trades |
getEarningsSurprises(ticker) |
Historical beats/misses |
getAccelerationMatrix(...) |
Revenue & earnings acceleration |
getLargeShareholderFilings(ticker) |
13D/13G activist & ownership filings |
getLaborEfficiency(ticker) |
Revenue/employee, SGA/employee, headcount |
getETFExposure(ticker) |
Reverse ETF lookup with weights |
getClinicalTrials(ticker?, phase?) |
Biotech/pharma trial pipeline |
getLobbyingActivity(ticker?, sector?) |
Federal lobbying spend & issues |
Advanced Analysis (11 tools — Tier 2)
| Tool | Description |
|---|---|
getNextEarnings(ticker) |
Next earnings date and reporting details |
getPeerComparison(ticker) |
Peer group analysis with relative rankings |
getValuationHistory(ticker) |
Historical P/E, P/S, EV/EBITDA time series |
getRiskMetrics(ticker) |
Beta, max drawdown, Sharpe, Sortino, VaR |
getAnnualTrends(ticker) |
Multi-year revenue, income, margins trajectory |
getKPIs(ticker) |
Industry-specific KPIs and operational metrics |
getAttentionSignals(ticker) |
Wikipedia + Google Trends anomalies |
getAcademicFactors(ticker) |
Residual income, accruals quality, Fama-French |
getForesightScore(ticker) |
ML composite foresight/trajectory score |
getInsiderSignals(ticker) |
Derived insider conviction patterns and clusters |
getAIVerdict(ticker) |
AI analyst buy/hold/sell verdict with thesis |
Business Exclusive (13 tools — Tier 3)
| Tool | Description |
|---|---|
getGovtContracts(ticker?, agency?) |
Government contract awards (by company or screened) |
getVolatilityProfile(ticker) |
Volatility analysis: risk metrics, vs SPY benchmark |
getRedditMentions(ticker?) |
Reddit sentiment & mention trends |
getMergerDeals(status?, ticker?) |
Active M&A deals with spreads |
getBatchEarningsPredictions(tickers) |
ML predictions for multiple tickers |
getETFBreakdown(ticker) |
ETF holdings, sector breakdown |
getIPOCalendar(range?) |
Upcoming IPOs with pricing details |
getAttentionAlerts(days?) |
Unusual attention spikes across universe |
getMacroSentiment() |
Full sentiment: AAII, VIX, put/call, ISM |
getGovtContractSummary(ticker) |
Aggregated contract exposure: 12-month totals, top agencies |
getBatchRiskMetrics(tickers) |
Risk metrics for multiple tickers |
getFactorModelScores(model) |
18 academic factor models with scores |
getMarketStreaks(sector?) |
Stocks on quality improvement streaks |
Resources
| Resource URI | Description | Cache |
|---|---|---|
finstill://market/signal |
Current market signal | 5 min |
finstill://market/regime |
Macro regime | 1 hour |
finstill://market/fear-greed |
Fear & Greed Index | 1 hour |
finstill://sectors/overview |
Sector aggregates | 1 hour |
finstill://company/{ticker} |
Company summary | 24 hours |
finstill://meta/tools |
Tool documentation | Static |
finstill://meta/coverage |
Covered tickers | 24 hours |
Prompts
| Prompt | Args | Description |
|---|---|---|
analyze-stock |
{ticker} |
Full stock analysis (15 tools) |
compare-stocks |
{tickers} |
Side-by-side comparison |
screen-stocks |
{criteria} |
NL criteria → stock screen |
macro-outlook |
— | Macro environment summary (10 tools) |
sector-rotation |
— | Sector rotation analysis |
earnings-preview |
{ticker} |
Pre-earnings analysis (10 tools) |
portfolio-health-check |
{tickers} |
Portfolio health: quality, risk, concentration |
alternative-data-scan |
{ticker} |
Non-traditional data deep dive |
market-dashboard |
— | Comprehensive market overview (13 tools) |
deep-value-hunt |
{sector?} |
Deep value stock discovery |
Programmatic Usage
import { createServer, connectStdio } from '@fffinstill/mcp-server';
const server = createServer({
apiUrl: 'https://fffinstill.com',
apiKey: process.env.FINSTILL_API_KEY!,
});
await connectStdio(server);
MCP Compliance
This server implements the full MCP specification:
- Tool annotations: All 69 tools declare
readOnlyHint: true,destructiveHint: false,idempotentHint: true,openWorldHint: true, and human-readabletitlefields - Resource metadata: All resources include
title,mimeType,description, and audience/priority annotations - Capabilities: Explicitly declares
tools,resources, andpromptswithlistChanged: false - Server instructions: Provides server-level instructions for LLM context
Rate Limits
Rate limits depend on your subscription tier:
| Plan | Requests/min | Monthly cap |
|---|---|---|
| Founding | 20 | 1,000 |
| Pro | 20 | 1,000 |
| Business | 60 | Unlimited |
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。