Back to Browse

Quant Brain MCP Server

Developer ToolsUse Caution3.2MCP RegistryLocalRemote
Free

Server data from the Official MCP Registry

Stock analysis MCP: 38 indicators, 7 optimizers, 8 backtests, US + India sectors.

About

Stock analysis MCP: 38 indicators, 7 optimizers, 8 backtests, US + India sectors.

Remote endpoints: streamable-http: https://mcp-quant-brain.onrender.com/mcp

Security Report

3.2
Use Caution3.2High Risk

This MCP server implements quantitative financial analysis with reasonable security fundamentals, but has several concerns that prevent a higher score. Missing authentication on the metrics endpoint, unvalidated user inputs in ticker resolution and optimization parameters, overly broad exception handling, and sensitive data exposure through telemetry create moderate-risk vulnerabilities. The code quality issues and lack of input sanitization in financial data handling require user awareness. Supply chain analysis found 10 known vulnerabilities in dependencies (1 critical, 9 high severity).

4 files analyzed · 20 issues found

Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.

Permissions Required

This plugin requests these system permissions. Most are normal for its category.

HTTP Network Access

Connects to external APIs or services over the internet.

env_vars

Check that this permission is expected for this type of plugin.

database

Check that this permission is expected for this type of plugin.

File System Read

Reads files on your machine. Normal for tools that analyze or process local data.

How to Install & Connect

Available as Local & Remote

This plugin can run on your machine or connect to a hosted endpoint. during install.

Documentation

View on GitHub

From the project's GitHub README.

mcp-quant-brain

MCP server for stock analysis, strategy backtesting, portfolio optimization, and chart generation.

38 curated quantitative indicators • 7 portfolio optimizers • 8 backtesting workflows • U.S. + Indian equities • production usage telemetry

This product is built for users who want to ask natural-language questions like:

  • "Analyze Indian sectors and tell me the best sector by risk-adjusted score."
  • "Backtest MACD on IREDA.NS and show charts."
  • "Optimize a US-India portfolio and explain risk."

Product Use Cases

  • Sector rotation analysis: identify the best-performing sector for a user-defined timeframe (for example 6m or 1yr) using return, volatility, drawdown, momentum, and correlation.
  • Portfolio construction: optimize allocations across US and India tickers using methods like MVO, HRP, max Sharpe, min volatility, CVaR, and semivariance.
  • Strategy validation: run rule-based backtests (MACD, RSI mean reversion, SMA crossover, breakout) before making discretionary decisions.
  • Chart-first review: generate chart packs for portfolio diagnostics, strategy behavior, and sector-level risk structure.
  • Company context enrichment: combine technical outputs with company-profile metadata for better explainability.

What You Get

  • US and India ticker support (AAPL, RELIANCE.NS, etc.)
  • 38 curated quantitative indicators exposed through 6 grouped analysis tools
  • Strategy backtests (MACD, RSI mean reversion, SMA crossover, breakout)
  • Portfolio optimization (MVO, HRP, max Sharpe, min volatility, Black-Litterman, CVaR, semivariance)
  • Sector intelligence (returns, volatility, momentum, drawdown, correlation, best-sector selection)
  • Chart pack generation with isolated chart configs and default chart-first layout

Use the Hosted MCP URL

Use streamable HTTP transport and connect your MCP client to:

  • https://mcp-quant-brain.onrender.com/mcp

Free-tier Render note:

  • The server may take up to 1-2 minutes to spin up if it was sleeping.
  • If the first request times out, wait and retry.

Connect in Claude (Desktop or Web)

Use connectors to add this MCP product in Claude.

Claude Desktop

  1. Open Claude Desktop settings.
  2. Go to Connectors or MCP integrations.
  3. Add a new connector.
  4. Choose Streamable HTTP transport.
  5. Set the connector URL to https://mcp-quant-brain.onrender.com/mcp.
  6. Save and connect.
  7. When Claude asks for tool permissions, click Accept All so all analysis and chart tools are available.

Claude Web

  1. Open Claude in browser and go to settings.
  2. Open Connectors.
  3. Add a custom connector.
  4. Select Streamable HTTP transport.
  5. Use https://mcp-quant-brain.onrender.com/mcp as the connector URL.
  6. Complete connection.
  7. When prompted for tool permissions, click Accept All.

If your first connection attempt fails, wait up to 2 minutes and retry once (free-tier cold start).

Core Tools You’ll Use

Indicator analysis (6 grouped tools, 38 indicators)

Each tool runs its whole group by default, or a subset via the optional indicators argument (for example analyze_momentum(ticker="AAPL", indicators=["rsi", "macd"])).

ToolIndicators
analyze_momentumrsi, macd, roc, cci, stoch, stochrsi, tsi, willr
analyze_technical_levelssma, ema, hma, kama, ichimoku, supertrend, vwap, vwma
analyze_trendadx, aroon, chop, psar, vortex, zigzag
analyze_volatilityatr, bbands, donchian, kc, stdev, ui
analyze_volumeobv, cmf, mfi, ad, pvt
analyze_statisticslog_return, zscore, skew, kurtosis, entropy

Portfolio, backtesting, and reporting

  • generate_optimized_verdict
  • generate_chart_pack
  • generate_charts
  • plot_charts
  • get_company_profile
  • find_sector_stock_pipeline_tool
  • analyze_sector_intelligence_tool
  • Strategy tools like backtest_macd_momentum, backtest_rsi_mean_reversion

Example Questions (NLP)

  • "Analyze IREDA.NS over 1 year, run MACD momentum backtest, and show charts."
  • "Compare IT, Bank, Auto, Metal, Pharma, Realty sectors by return, risk, drawdown, and momentum."
  • "Give me the best Indian sector using risk-adjusted ranking and show sector correlation matrix."
  • "Optimize AAPL, MSFT, NVDA, RELIANCE.NS with CVaR and summarize risk flags."

Render Endpoints

  • MCP endpoint: https://mcp-quant-brain.onrender.com/mcp
  • Health endpoint: https://mcp-quant-brain.onrender.com/health
  • Metrics endpoint: https://mcp-quant-brain.onrender.com/metrics/summary (bearer token required)

Usage Telemetry

Every MCP tool call records tool_name, tool_category, session_id, duration_ms, and success to Postgres. Writes are fire-and-forget on a background thread, so telemetry never blocks or fails a tool call. With no DATABASE_URL configured, telemetry is a silent no-op and the server runs normally.

Environment variables

VariablePurpose
DATABASE_URLPostgres/Supabase connection URI (use the Supabase session pooler URI). Unset disables telemetry.
METRICS_TOKENBearer token protecting /metrics/summary. Unset returns 503.

The mcp_tool_events table and its indexes are created automatically on first connection — no manual migration needed.

Reading your metrics

curl -H "Authorization: Bearer $METRICS_TOKEN" \
  https://mcp-quant-brain.onrender.com/metrics/summary

Returns total_requests, unique_sessions, backtests, optimizations, success_rate, p50_latency, and p95_latency.

Troubleshooting

"Missing session ID"

The server is configured with stateless HTTP mode. If you still see this:

  1. Reconnect MCP client.
  2. Ensure URL is exactly /mcp.

"MCP server connection lost" / 504 / timeout

Usually cold-start or proxy timeout on free tier:

  1. Wait up to 2 minutes.
  2. Retry once.
  3. Check /health.

Charts not visible

Use generate_charts or plot_charts for image-friendly responses.

Notes

  • This is not financial advice.
  • Data quality depends on upstream sources (mainly Yahoo Finance via yfinance).
  • Use proper ticker suffixes for Indian stocks (for example .NS).

Reviews

No reviews yet

Be the first to review this server!