esolat-mcp 1.0.5

MCP server for accurate Malaysian prayer times (official JAKIM/e-Solat), nearest masjid and surau, and Islamic calendar events.

Gives AI assistants like Claude Desktop and Claude Code reliable, hallucination-free access to Islamic worship tools with smart Malaysia-first routing.

Features

• Official JAKIM data for Malaysia (via Waktusolat.app API)
• Dhuha time automatically calculated (+28 minutes after Syuruk)
• Malay Hijri month names (Muharram, Safar, etc.)
• Global fallback using Aladhan API
• Nearest mosques/suraus with Google Maps + Waze deep links
• Yearly Islamic events (Eid, etc.)
• Dual mode: Local (stdio) + Remote (Docker + secure HTTP)
• Health dashboard for upstream APIs

Tools

The server registers three powerful tools that Claude can discover and call automatically:

1. get_monthly_prayer_times

   • Get full monthly prayer schedule (Fajr, Syuruk, Dhuha, Dhuhr, Asr, Maghrib, Isha)
   • Accepts place name or latitude/longitude
   • Returns both Gregorian and Hijri dates

2. find_nearest_mosques

   • Find up to 15 nearest masjid/surau
   • Default radius: 5 km (configurable)
   • Malaysia uses official e-Solat data; global uses OpenStreetMap
   • Includes distance, coordinates, Google Maps & Waze links

3. get_yearly_islamic_events

   • Major Islamic dates and holidays for a given year
   • Malaysia-aware routing

Quick Start

Option 1: Local / Stdio Mode (Recommended for Claude Desktop)

No clone needed:

uvx esolat-mcp

Claude Desktop config example:

{
  "mcpServers": {
     "esolat": {
      "command": "uvx",
      "args": ["esolat-mcp"]
    }
  }
}

Option 2: Docker / Remote HTTP Mode (Self-hosted)

1. Clone the repo and set up token:

git clone https://github.com/zubir2k/esolat-mcp.git
cd esolat-mcp
cp .env.example .env

2. Generate a strong token:

python -c "import secrets; print(secrets.token_urlsafe(32))"
# or: openssl rand -hex 32

3. Paste it into .env as MCP_WEBHOOK_TOKEN=your_token_here

4. Build and Start the container:

docker build -t esolat-mcp:latest .
docker compose up -d

5. Endpoints

- Streamable MCP HTTP Endpoint (main one for AI clients):

http://your-server-ip:8626/api/webhook/<MCP_WEBHOOK_TOKEN>/mcp

- Health Dashboard (useful for monitoring):

http://your-server-ip:8626/api/webhook/<MCP_WEBHOOK_TOKEN>

(GET request without /mcp — shows status of JAKIM, Aladhan, and Overpass APIs)

Usage Examples (Claude)

Prayer Times

"What are today's prayer times in Kuala Lumpur?"
"Show me full prayer schedule for Penang this Ramadan."
"Prayer times for Kota Kinabalu next week."

Mosques

"Find the nearest mosque to me."
"Mosques within 10km of Petaling Jaya."
"Nearest surau from KLCC."

Events & Planning

"Major Islamic events in 2026 for Malaysia."
"Plan my trip to Johor Bahru: prayer times + nearest mosques + Eid dates."

Visual

Claude Config

Claude Prompt

Credits

• e-solat JAKIM - For the official prayer times
• WaktuSolat.app - Prayer Time by GPS
• Model Context Protocol - For the MCP framework

License

This project is licensed under the MIT License.

Disclaimer & Data Source

Important Notice & Reliability Disclaimer

This integration pulls data directly from the official e-Solat JAKIM (Department of Islamic Development Malaysia) portal. However, please note:

• No Liability: This integration is a community-driven project provided "as is" without any guarantees. The maintainer is not solely or legally responsible for any discrepancies, inaccuracies, delays, or omissions in prayer times or calendar dates.
• Verify Important Times: Users are strictly advised to regularly check and verify times against the official JAKIM eSolat Portal or official local announcements, especially for critical obligations (e.g., fasting, community prayers).
• Network & Upstream Dependencies: Synchronization depends on upstream API availability and local network connectivity. Discrepancies caused by unexpected server updates from JAKIM or local server downtime are outside the control of this software.

By using this tool, you acknowledge and agree that the developer holds no liability for missed schedules or data inaccuracies.