Islamic Finance MCP Server: Connect AI Agents to Shariah Data

The Model Context Protocol (MCP) is an open standard that lets AI assistants (Claude, Cursor, and others) connect to external data sources and tools. The Halal Terminal MCP server gives AI agents direct access to Shariah screening, ETF analysis, zakat calculators, and more — 22 tools and 7 resources that make AI assistants Islamic finance-aware.

What Does This Mean?

Instead of asking an AI "Is NVIDIA halal?" and getting a generic answer, you can connect the Halal Terminal MCP server and get real-time screening data — actual financial ratios, methodology-by-methodology verdicts, and purification rates — directly in your AI conversation.

Available MCP Tools

Category	Tools	Description
Screening	`screen_stock`, `get_result`, `get_all_results`	Screen any stock against 5 Shariah methodologies
Bulk Screening	`screen_bulk`, `bulk_status`, `bulk_results`, `bulk_summary`	Screen entire indices asynchronously
ETF Analysis	`screen_etf`, `etf_holdings`, `etf_info`, `compare_etfs`	Screen ETFs per-holding and compare side-by-side
Zakat & Purification	`calculate_zakat`, `calculate_purification`, `dividend_purification`	Calculate zakat and purification amounts
Database	`search_stocks`, `stock_info`, `database_stats`, `suggest`	Search and browse the stock database
Market Data	`dividends`, `news`, `sec_filings`	Dividend history, news, and SEC filings
System	`health`, `token_costs`	Check API health and token pricing

Setup: Claude Desktop

Add this to your Claude Desktop claude_desktop_config.json:

{
  "mcpServers": {
    "halal-terminal": {
      "command": "npx",
      "args": ["-y", "@anthropic/mcp-remote", "https://api.halalterminal.com/mcp/sse"],
      "env": {
        "API_KEY": "YOUR_HALAL_TERMINAL_API_KEY"
      }
    }
  }
}

Setup: Cursor IDE

In Cursor Settings > MCP Servers, add:

{
  "halal-terminal": {
    "url": "https://api.halalterminal.com/mcp/sse",
    "headers": {
      "X-API-Key": "YOUR_KEY"
    }
  }
}

Example Conversations

Once connected, you can ask your AI assistant natural language questions and get data-backed answers:

"Is Tesla halal?"

The AI calls screen_stock("TSLA") and returns:

Tesla (TSLA) is Shariah-compliant across all 5 methodologies:
- AAOIFI: PASS (debt/assets 6.8% < 30%)
- DJIM: PASS (debt/market-cap 0.9% < 33%)
- FTSE: PASS
- MSCI: PASS
- S&P: PASS
Purification rate: 1.5%

"Compare SPUS and HLAL"

The AI calls compare_etfs(["SPUS", "HLAL"]) and returns a detailed side-by-side comparison with compliance rates, holdings overlap, expense ratios, and purification rates.

"Calculate zakat on my portfolio: 100 shares of AAPL, 50 shares of MSFT"

The AI calls calculate_zakat with your holdings and returns the nisab threshold, whether you're above it, and the exact zakat amount per holding.

Building with MCP

If you're building an AI application that needs Islamic finance awareness, the MCP server lets you add this capability without building screening logic from scratch. Use cases:

AI financial advisors that automatically check Shariah compliance before recommending stocks
Chatbots for Islamic banks that can answer customer questions about fund compliance
Research assistants that can screen portfolios and calculate zakat in conversation
Investment platforms that use AI to explain screening results in natural language

Build with Halal Terminal

Halal Terminal API

58+ endpoints, 5 screening methodologies, ETF analysis, zakat calculators, and MCP tools. Free tier available.

Explore the API → Try the Terminal instead

Key Takeaways

MCP lets AI assistants access real-time Shariah data instead of guessing
22 tools covering screening, ETFs, zakat, purification, and market data
Works with Claude Desktop, Cursor, and any MCP-compatible client
No Islamic finance platform other than Halal Terminal offers MCP integration
Setup takes 2 minutes — just add the config and your API key