Sentimatix MCP Server
Real-time Indian stock market intelligence via the Model Context Protocol (MCP). Plug Sentimatix directly into Claude Desktop, Cursor, or any MCP-compatible AI agent to get live NSE/BSE sentiment, news, and technical analysis.
🏗️ Architecture
User Query → Claude/AI Agent → MCP Server → Tools
↓
┌─────────────────────┐
│ News Sentiment API │ ← Sentimatix /api/v1/news
│ Stock Price API │ ← Sentimatix /api/v1/stocks
│ Sector Sentiment │ ← Sentimatix /api/v1/sentiment
│ RAG Evidence │ ← Supabase Vector DB
└─────────────────────┘
↓
Synthesized Market Intelligence
🛠️ Available Tools (10 total)
| Tool | Description | Example |
|---|---|---|
explain_price_change |
Orchestrator — explains why a stock moved using price, news + sentiment | (RELIANCE, 7 days) |
analyze_stock_enhanced |
Deep single-stock research report | (HDFCBANK, detailed) |
compare_stocks |
Side-by-side comparison of two NSE stocks | (TCS vs INFY) |
get_stock_summary |
Price metrics: change%, high, low, volume | (TATAMOTORS, 7 days) |
get_historical_prices |
Daily OHLCV time-series data | (SBIN, 2024-01-01, 2024-03-01) |
get_news_sentiment |
News articles with NLP sentiment scores | (ZOMATO, last 7 days) |
get_sentiment_aggregate |
Aggregated sentiment stats for a period | (ADANIENT, 30 days) |
get_technical_analysis |
RSI, MACD, Bollinger Bands, support/resistance | (ITC, 90 days) |
calculate_correlation |
Pearson correlation between two NSE stocks | (RELIANCE, TCS) |
get_rag_evidence |
Semantic search over the news corpus | (INFY, "earnings beat") |
🚀 Quick Start (Claude Desktop)
Add this to your claude_desktop_config.json:
{
"mcpServers": {
"sentimatix": {
"command": "python",
"args": ["path/to/apps/mcp/mcp_stdio.py"],
"env": {
"BACKEND_API_URL": "https://sentimatix-production.up.railway.app/api"
}
}
}
}
No API key is required for the default free tier.
📊 Example Query Flow
User: "Why did Reliance drop in the last 7 days?"
MCP Workflow:
- ✅
get_stock_summary("RELIANCE", 7)→-3.1% decline - ✅
get_news_sentiment("RELIANCE", ...)→12 news items, avg sentiment: -0.4 - ✅
get_rag_evidence("RELIANCE", "price drop reasons")→5 relevant articles - ✅
get_technical_analysis("RELIANCE")→RSI: 38 (Oversold), MACD: Bearish - 🤖 LLM Synthesis:
RELIANCE declined 3.1% (₹2,850 → ₹2,762) over the last 7 days.
Top Reasons (with evidence):
1. Weak Q4 refining margins — reported by Economic Times (sentiment: -0.5)
2. Sector-wide FII selling in Oil & Gas — MoneyControl (sentiment: -0.3)
3. RSI at 38 confirms oversold territory; key support at ₹2,740
Confidence: HIGH
🔧 Environment Variables
BACKEND_API_URL=https://sentimatix-production.up.railway.app/api
SUPABASE_URL=https://your-project.supabase.co # For RAG features
SUPABASE_KEY=your_anon_key # For RAG features
OPENAI_API_KEY=your_openai_key # For AI synthesis
📚 Full Documentation
- Connection Guide — Claude Desktop, Cursor setup
- Quick Start — Step-by-step installation
- REST API Docs — Full OpenAPI reference
Metadata
Release files for sentimatix-mcp 1.0.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| sentimatix_mcp-1.0.2.tar.gz | 56.7 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| sentimatix_mcp-1.0.2-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 121.0 kB
Release files / sentimatix_mcp-1.0.2.tar.gz
| Download URL | sentimatix_mcp-1.0.2.tar.gz |
|---|---|
| Size | 56.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
0e4591f36ad69d9886990e00c2968782fafbd7f681fa94f9aaa45174ae263aab
|
|
BLAKE2b-256 checksum How to use checksums |
95c411b00d62797ac169a12c3f5b0ac0bc47319e3fc5b05c8a2bbbaddb0e3cf0
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.2
|
Release files / sentimatix_mcp-1.0.2-py3-none-any.whl
| Download URL | sentimatix_mcp-1.0.2-py3-none-any.whl |
|---|---|
| Size | 64.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
194c2e7fe5c6faaf5ed121b6bb4809d1b2a3dd6877fd517df88783d4304ff175
|
|
BLAKE2b-256 checksum How to use checksums |
c7fcd777cf295ba10aae451008e2f5bd98e56c482684a48d925def5971f39e8a
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.2
|