
ManaMurah MCP Server
An AI-optimized Model Context Protocol server for querying Malaysian consumer goods prices from official KPDN Pricecatcher data, enabling natural language price searches and comparisons across regions.
README
ManaMurah MCP Server
A Model Context Protocol (MCP) server for Malaysian price data from KPDN Pricecatcher. This server enables direct integration with Claude Desktop and other MCP-compatible AI tools for querying Malaysian consumer goods prices.
Features
🇲🇾 Official Malaysian Price Data - KPDN Pricecatcher data via OpenDOSM
🤖 AI-Optimized - Natural language queries with intelligent parsing
⚡ Serverless - Deployed on Cloudflare Workers for global performance
🔒 Rate Limited - Built-in abuse protection and fair usage
📊 Rich Analytics - Price comparisons, trends, and market insights
🎯 Claude Desktop Ready - One-click setup for Claude Desktop integration
Live Demo
The MCP server is deployed and accessible at:
- Production: https://mcp.manamurah.com
- API Endpoint: https://mcp.manamurah.com/mcp
- Status: https://mcp.manamurah.com/ (returns server info)
Quick Start
Deploy to Cloudflare Workers
# Clone or create from template
npm create cloudflare@latest manamurah-mcp-server --template=cloudflare/ai/demos/remote-mcp-authless
# Replace src/ contents with ManaMurah implementation
# (Copy files from this directory)
# Install dependencies
npm install
# Deploy to Cloudflare Workers
npm run deploy
Connect to Claude Desktop
- Get your deployed Workers URL (e.g.,
https://mcp.manamurah.com
) - Add to Claude Desktop MCP configuration:
{
"manamurah": {
"command": "node",
"args": ["/path/to/mcp-client.js"],
"env": {
"MCP_SERVER_URL": "https://mcp.manamurah.com/sse"
}
}
}
- Restart Claude Desktop
- Start asking about Malaysian prices!
Available Tools
🔍 get_malaysian_prices
Search for current prices with natural language queries.
Examples:
- "rice prices in Kuala Lumpur"
- "cheapest chicken in Penang hypermarkets"
- "cooking oil under RM20 in Selangor"
📊 compare_prices
Compare prices across different regions or retail chains.
Examples:
- Compare rice prices between KL and Penang
- Find price differences across retail chains
- Regional price analysis for specific items
📈 analyze_price_trends
Analyze price trends and market patterns over time.
Examples:
- Price volatility analysis
- Seasonal price patterns
- Regional market trends
💡 get_market_insights
Get market intelligence and insights about price anomalies.
Examples:
- Recent price changes
- Market anomaly detection
- Regional price differences
Example Usage
Basic Price Search
User: "What are rice prices in Kuala Lumpur?"
MCP Response:
📊 Summary: Rice prices in Kuala Lumpur range from RM15.20-RM25.80 per 5kg
💡 Key Insights:
• Hypermarkets offer 18% lower prices than convenience stores
• Significant price variation exists across different retailers
📈 Price Statistics:
• Average: RM18.50
• Range: RM15.20 - RM25.80
• Price Variation: 32%
[Detailed price listings follow...]
Price Comparison
User: "Compare chicken prices between Penang and Johor"
MCP Response:
📊 Summary: Penang has lower average chicken prices (RM8.20) compared to Johor (RM9.10)
## Regional Comparison
### 1. Penang
• Average Price: RM8.20
• Price Range: RM7.50 - RM9.80
• Sample Size: 15 price points
### 2. Johor
• Average Price: RM9.10
• Price Range: RM8.20 - RM11.50
• Sample Size: 12 price points
💡 Comparison Insights:
• Most Affordable: Penang (RM8.20 average)
• Potential Savings: RM0.90 (9.9%) by choosing Penang
Development
Local Development
# Install dependencies
npm install
# Start development server
npm run dev
# Type checking
npm run type-check
# Linting
npm run lint
Project Structure
src/
├── index.ts # Main MCP server implementation
├── utils/
│ ├── api-client.ts # ManaMurah API client
│ ├── query-parser.ts # Natural language query parsing
│ ├── response-formatter.ts # MCP response formatting
│ └── rate-limiter.ts # Rate limiting implementation
└── types/
└── manamurah.ts # TypeScript type definitions
Configuration
Environment variables in wrangler.toml
:
[vars]
MANAMURAH_API_BASE = "https://api.manamurah.com"
RATE_LIMIT_ENABLED = "true"
CACHE_TTL = "300"
MAX_QUERIES_PER_MINUTE = "10"
MAX_QUERIES_PER_HOUR = "100"
Rate Limits
- Per Minute: 10 requests
- Per Hour: 100 requests
- Automatic Cleanup: Old request data is cleaned up automatically
Rate limits help ensure fair usage and prevent abuse while allowing genuine research and analysis.
Features
Natural Language Processing
- Intelligent extraction of items, locations, and price constraints
- Support for Malaysian terms (e.g., "beras" for rice, "ayam" for chicken)
- Price range detection ("under RM20", "between RM10 and RM15")
- Location recognition for all Malaysian states and major cities
Rich Response Formatting
- Markdown-formatted responses optimized for Claude Desktop
- Statistical analysis with averages, ranges, and insights
- Suggested follow-up questions for continued exploration
- Data source attribution and freshness indicators
Error Handling
- User-friendly error messages with helpful suggestions
- Graceful degradation when data is unavailable
- Query improvement recommendations
- Comprehensive error logging for debugging
Data Source
Official Government Data: KPDN Pricecatcher program via OpenDOSM
- Daily data updates (subject to government publication schedules)
- Comprehensive coverage of Malaysian retail prices
- Data includes hypermarkets, supermarkets, convenience stores, and grocery shops
- Covers all Malaysian states and major urban centers
Support
Getting Help
- Documentation: api.manamurah.com/docs
- AI Integration Guide: Complete guide for AI developers
- Issues: GitHub Issues
Contact
- General Support: support@manamurah.com
- AI Integration: ai-support@manamurah.com
- Enterprise: enterprise@manamurah.com
License
MIT License - see LICENSE file for details.
Contributing
Contributions welcome! Please read our contributing guidelines and submit pull requests for any improvements.
Built with ❤️ for the Malaysian data community
Making Malaysian price data accessible to AI tools and researchers worldwide.
推荐服务器

Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。