Daniel Rosehill

JLM Shelters: A Website To Find Public Bomb Shelters In Jerusalem

Israel Jerusalem shelters emergency preparedness open source
JLM Shelters: A Website To Find Public Bomb Shelters In Jerusalem

I built jlmshelters.com to help people in Jerusalem quickly find their nearest public bomb shelter (miklat tziburi). The site currently maps 198 verified shelters across the city.

Why this exists

When you have 90 seconds to reach a shelter, you need to know exactly where your nearest one is. The official municipal data exists but isn't always easy to access or search through quickly. This site makes it simple: search by address or neighbourhood, filter by shelter type and accessibility, and get a one-tap navigation link to Google Maps or Waze.

Features

  • 198 public shelters in Jerusalem mapped and searchable

  • Free-text search across names, addresses, and neighbourhoods

  • Find nearest shelters by your current GPS coordinates

  • Filter by shelter type, capacity, and accessibility

  • One-tap navigation links to Google Maps or Waze

  • Neighbourhood breakdowns with shelter counts

MCP integration for AI agents

The site is backed by the Miklat MCP server, which runs on Cloudflare Workers and exposes the same shelter data as MCP tools. This means AI assistants like Claude can help users find shelters conversationally. The MCP endpoint is globally accessible at https://mcp.jlmshelters.com/mcp.

danielrosehill/Miklat-MCP ★ 4

MCP to enable AI agents to guide users towards miklatim tziburim (public shelters) in Israel

TypeScriptUpdated Mar 2026
mcppreparednesssafety

Data and limitations

Shelter data is community-maintained in the Miklat-MCP-Data repository using GeoJSON format with a three-stage validation pipeline. Currently covers Jerusalem only. Contributions for other Israeli cities are welcome.

danielrosehill/Miklat-MCP-Data ★ 0

Data pipeline for Miklat MCP

Updated Mar 2026

Important: this is an informational tool, not an official resource. Always verify shelter locations with municipal sources and follow Home Front Command instructions during emergencies.

Daniel Rosehill

Daniel Rosehill

AI developer and technologist specializing in AI systems, workflow orchestration, and automation. Specific interests include agentic AI, workflows, MCP, STT and ASR, and multimodal AI.