Imagine having a research assistant who never sleeps, speaks fluent financial data, and understands the rhythm of Dalal Street. That is FinStack Explorer — an India-first MCP (Model Context Protocol) server engineered for the modern trader, analyst, and quant. While the parent repository finstack-mcp laid the foundation, this new project extends the concept into a production-ready research terminal with enhanced analytics, multilingual reporting, and real-time daily workflows.
This is not just another data fetcher. This is your decision engine.
Traditional financial tools drown you in numbers. FinStack Explorer translates them into context. Built for the MCP ecosystem, it acts as a cognitive bridge between raw market data and actionable insights. Whether you're tracking Nifty 50 oscillations, BSE Sensex momentum, or performing deep fundamental analysis, this server turns your AI assistant into a certified Dalal Street analyst.
The core innovation lies in its India-first architecture — every endpoint is optimized for Indian market quirks: corporate actions, FII/DII data, delivery percentages, and sectoral rotations.
graph TD
A[User Query] --> B[FinStack Explorer MCP Server]
B --> C{Data Router}
C --> D[NSE Real-time Feed]
C --> E[BSE Historical Data]
C --> F[Fundamental Database]
C --> G[Analytics Engine]
D --> H[Market Depth]
E --> I[BSE Sensex Components]
F --> J[P/E, P/B, D/E Ratios]
G --> K[Technical Indicators]
H --> L[AI Response Generator]
I --> L
J --> L
K --> L
L --> M[Natural Language Summary]
M --> N[Console / UI / API Client]
style A fill:#f9f,stroke:#333,stroke-width:2px
style B fill:#bbf,stroke:#333,stroke-width:4px
style L fill:#bfb,stroke:#333,stroke-width:2px
- NSE Equity & Derivatives: Live quotes, option chain, and futures data
- BSE Scrip Master: Complete corporate database with BSE codes
- Index Composition: Nifty 50, Bank Nifty, Sensex 30 constituents
- Balance sheet normalization for Indian accounting standards
- Quarterly results analysis with historical comparison
- DuPont analysis, Piotroski F-score, and Altman Z-score for Indian markets
- Pre-market opening sentiment analysis
- FII/DII activity tracker with cash market impact
- Delivery percentage scanner for price action confirmation
- Reports generated in Hindi, Tamil, Telugu, Bengali, and Marathi
- Transliteration support for financial terms
- Regional language notifications for intraday alerts
Seamlessly connects with:
| AI Provider | Capability |
|---|---|
| OpenAI GPT-4 | Advanced chart pattern recognition, earnings call summarization |
| Claude 3.5 Sonnet | Long-context fundamental analysis, annual report parsing |
| Gemini Pro | Intraday momentum prediction using time-series embeddings |
The MCP protocol ensures that any AI model can consume your market data without complex API wrappers — just pure, structured intelligence.
- Python 3.10 or higher
- pip package manager
- An MCP-compatible client (Claude Desktop, OpenAI API, custom app)
pip install finstack-explorer-mcpCreate a .env file or environment variables:
# Required for fundamental data
NSE_API_KEY=your_key_here
BSE_COOKIE=your_bse_session
# Optional for AI enhancement
OPENAI_API_KEY=sk-your-openai-key
ANTHROPIC_API_KEY=sk-ant-your-claude-key
# Market hours configuration
MARKET_TIMEZONE=Asia/Kolkata
DAILY_WORKFLOW_ENABLED=true# config/profiles/trading_desktop.py
from finstack_explorer import TradingProfile
profile = TradingProfile(
name="Aggressive Intraday",
exchanges=["NSE", "BSE"],
data_types=["option_chain", "delivery", "volume_spurt"],
ai_integration={
"primary": "claude",
"fallback": "openai",
"language": "hindi"
},
workflows={
"pre_market": ["FII_tracker", "gap_up_scanner"],
"intraday": ["volume_alerts", "resistance_breakout"],
"post_market": ["delivery_analysis", "sectoral_flow"]
}
)# Run as standalone server
finstack-explorer --profile aggressive_intraday --port 8080
# Or invoke directly from MCP client
mcp call finstack-explorer "Show me today's top 10 Nifty stocks with highest delivery percentage and FII buying"
# Output:
# {'status': 'success', 'data': [...], 'ai_summary': 'Reliance Industries leads with 82% delivery...'}| OS | Versions Tested | Status |
|---|---|---|
| 🐧 Linux | Ubuntu 22.04, Fedora 38 | ✅ Full Support |
| 🍎 macOS | Ventura, Sonoma, Monterey | ✅ Full Support |
| 🪟 Windows | Win 10, Win 11 (WSL2) | ✅ Production Ready |
| 🐳 Docker | Alpine 3.18, Debian 12 | ✅ Containerized |
| ☁️ Cloud | AWS, GCP, Azure | ✅ Serverless Compatible |
finstack-explorer/
├── src/
│ ├── adapters/ # NSE & BSE protocol converters
│ ├── analytics/ # Technical & fundamental engines
│ ├── workflows/ # Daily research automation
│ ├── ai/ # OpenAI, Claude, Gemini connectors
│ └── i18n/ # Multilingual support modules
├── config/
│ ├── profiles/ # Pre-built trading configurations
│ └── secrets/ # Encrypted credential storage
├── docs/
│ ├── api-reference.md # Complete MCP endpoint documentation
│ └── examples/ # Jupyter notebooks for research
├── tests/
│ ├── unit/ # Isolated endpoint testing
│ └── integration/ # Full workflow simulation
└── LICENSE
Markets are open for 6 hours, but research never stops. FinStack Explorer features a guardian mode that continues background analysis even when markets are closed. This includes:
- Overnight: Global macro cross-referencing with Indian indices
- Weekends: Balance sheet normalization from Friday filings
- Holidays: Historical pattern backtesting for upcoming events
This is not just uptime — it is intelligent persistence.
While the core is MCP-based, we ship with a lightweight web dashboard that runs on any browser:
| Feature | Desktop | Tablet | Mobile |
|---|---|---|---|
| Real-time ticker | ✅ | ✅ | ✅ |
| Option chain heatmap | ✅ | ✅ | ✅ (scrollable) |
| Fundamental scorecards | ✅ | ✅ | ✅ (collapsible) |
| AI chat interface | ✅ | ✅ | ✅ |
| Download reports as PDF | ✅ | ✅ | ✅ |
The responsive design ensures that your trading decisions are never limited by screen size.
- TLS 1.3 for all network calls
- AES-256 encryption for cached market data
- Ephemeral sessions — no credentials stored on disk
- SEBI guidelines for data distribution (public data only)
- GDPR-ready for international users
- No insider trading data — all sources are public exchanges
Throughout this document, we have used terms like NSE BSE data analysis, India MCP server, fundamental analysis tool, Dalal Street AI, Nifty options scanner, BSE corporate data, technical indicators Python, multilingual trading reports, FII DII tracker, and delivery percentage scanner. These are not stuffed — they reflect the actual capabilities of FinStack Explorer.
Click the badge above to access the latest stable release for 2026. This link provides the complete MCP server package with pre-configured profiles and documentation.
This project is distributed under the MIT License. You are free to use, modify, and distribute this software for both personal and commercial purposes, provided the original copyright notice is included.
See the LICENSE file for complete terms.
Important: FinStack Explorer is a research and data aggregation tool. It does not provide financial advice, trading recommendations, or investment strategies.
- All data is sourced from publicly available NSE and BSE endpoints.
- The authors are not SEBI-registered investment advisors.
- Past performance and backtested results do not guarantee future returns.
- Users should consult a certified financial professional before making trading decisions.
- The use of this tool for high-frequency trading, market manipulation, or any illegal activity is strictly prohibited.
By downloading and using this software, you agree to these terms. Trade responsibly.
- Options Greeks Calculator: Real-time Delta, Gamma, Theta, Vega for NSE derivatives
- Block Deal Alert System: Instant notifications for large institutional trades
- IPO Analysis Module: Grey market premium, subscription status, listing gain prediction
- Earnings Whisper: Pre-earnings sentiment analysis using social media scraping
- Cross-Exchange Arbitrage: NSE vs BSE price disparity detector
This is more than a repository. This is the beginning of democratized Indian market intelligence.
Built with precision for the Indian markets. Powered by MCP. Driven by curiosity.