israel-drugs

Israel Drugs MCP Server provides AI agents with access to Israel's Ministry of Health pharmaceutical data, enabling evidence-based medication guidance, safety assessments, and therapeutic recommendations. The server ships with three built-in information templates that structure data for clinical decision-making: compare_therapeutic_options, drug_safety_verification, and symptom_to_treatment_workflow. Use the Core Tools to search for drugs by name, explore therapeutic alternatives, verify safety information, and discover treatment pathways based on symptoms. When interacting with the MCP API, you can leverage the templates to present organized, cite-ready clinical information and to generate educational content tailored to the Israeli healthcare context.

Prerequisites:

• Node.js 18+
• npm (installed with Node.js)
• Internet access for real-time data

Installation steps:

1. Clone the repository: git clone https://github.com/DavidOsherDevDev/israel-drugs-mcp-server.git
2. Navigate to the project folder: cd israel-drugs-mcp-server
3. Install dependencies: npm install
4. Build the project (if required by the setup): npm run build
5. Start the server: npm start

Claude Desktop configuration (example): Add to Claude settings the following server entry (adjust paths as needed): { "mcpServers": { "israel-drugs": { "command": "node", "args": ["/path/to/israel-drugs-mcp-server/dist/index.js"], "env": {}, "disabled": false, "autoApprove": [] } } }

Notes and tips:

• The MCP server is Node.js-based and expects a compiled/built entry at dist/index.js; ensure you build before starting if your setup requires it.
• The mcp_config uses the server name israel-drugs. Ensure the command points to the correct entry file for your deployment.
• Environment variables can be used to configure licensing, data sources, or feature flags; you can populate env in the mcp_config accordingly.
• This server exposes real-time Israel-specific drug data including pricing, health basket status, and Ministry of Health approvals. Verify data sources if you deploy in a regulated environment.
• When integrating with Claude or other MCP clients, you can use the built-in templates to structure outputs for easy consumption by AI agents and end users.