Skip to main content
Glama
README.md
# Gulf Data MCP

An MCP server that gives Claude, ChatGPT, Cursor and other AI agents **live Gulf & Middle East data** plus **web search and web-to-Markdown** for grounding answers.

| Tool | What it returns |
|---|---|
| `dubai_property_transactions` | Official Dubai Land Department (DLD) registered sales, mortgages or gifts for any date range: count, value (AED), area, **price per sqm**, first sale vs resale, by community and property type. |
| `gulf_jobs` | Current jobs in the UAE, Saudi Arabia, Qatar, Kuwait, Bahrain or Oman from **LinkedIn, NaukriGulf, Bayt and GulfTalent** in one search, duplicates merged. |
| `gold_price` | Gold and silver in any currency (AED, INR, SAR, PKR, USD...), per gram, tola, kilo and 10 g, with 24k/22k/21k/18k/14k prices and optional history. |
| `amazon_best_sellers` | Best Sellers / New Releases / Movers & Shakers for **21 Amazon marketplaces** (UAE, Saudi, India, US, UK, Germany, Japan...). |
| `noon_products` | Product search on **Noon** (UAE, Saudi Arabia, Egypt): price, discount, rating, seller. |
| `web_search` | Web search results with each page's main text as clean Markdown, ready to cite. |
| `web_to_markdown` | Any URL (or a whole site) to clean, LLM-ready Markdown with token counts. |

Each tool runs a maintained, ready-made scraper on [Apify](https://apify.com/get_anything) using **your own Apify account**, so there are no shared rate limits and no data passes through a third-party server.

## Setup

1. Create a free Apify account and copy your API token from **Console → Settings → Integrations**.
2. Add the server to your MCP client.

**Claude Desktop / Cursor / any stdio client** (needs [uv](https://docs.astral.sh/uv/)):

```json
{
  "mcpServers": {
    "gulf-data": {
      "command": "uvx",
      "args": ["--from", "git+https://github.com/GetAnything-1/gulf-data-mcp", "gulf-data-mcp"],
      "env": { "APIFY_TOKEN": "your-apify-token" }
    }
  }
}
```

**Or with pip:**

```bash
pip install git+https://github.com/GetAnything-1/gulf-data-mcp
APIFY_TOKEN=your-apify-token gulf-data-mcp
```

Then ask your assistant things like:

- *"Which Dubai communities had the most apartment resales last month, and at what price per sqm?"*
- *"Find accountant jobs in Riyadh posted this week, grouped by company."*
- *"What's the 22k gold price per gram in Dubai and Mumbai right now?"*
- *"Show the top 10 Amazon.ae best sellers in electronics and compare with Noon prices."*
- *"Search the web for the latest UAE visa rule changes and summarise the top 5 sources with links."*

## Cost

Each call runs a pay-per-result scraper on your Apify account (typically $1–10 per 1,000 results; see each scraper's page). Apify's free plan includes monthly credit, which covers plenty of agent use. Keep `max_*` parameters small for quick questions.

| Tool | Scraper |
|---|---|
| `dubai_property_transactions` | [DLD Transactions Scraper](https://apify.com/get_anything/dld-transactions-scraper) |
| `gulf_jobs` | [GCC Jobs Aggregator](https://apify.com/get_anything/gcc-jobs-aggregator) |
| `gold_price` | [Gold Price Scraper](https://apify.com/get_anything/gold-price-scraper) |
| `amazon_best_sellers` | [Amazon Products & Best Sellers](https://apify.com/get_anything/amazon-ae-scraper) |
| `noon_products` | [Noon Scraper](https://apify.com/get_anything/noon-scraper) |
| `web_search` | [AI Web Search Scraper](https://apify.com/get_anything/ai-web-search-scraper) |
| `web_to_markdown` | [LLM Web Scraper](https://apify.com/get_anything/llm-web-scraper) |

## Notes

- Data comes from public pages and official public sources. No logins to the source sites.
- The scrapers don't solve CAPTCHAs or defeat bot protection. If a source refuses (Bayt and GulfTalent sometimes do from the cloud), those results are skipped and not charged.
- Live market examples built with this data: [gold price today](https://getanything-site.vercel.app/data/gold-price), [Dubai price per sqm](https://getanything-site.vercel.app/data/dubai-price-per-sqm).

## License

MIT. Questions or a data source you need? get.anything2026@gmail.com

TDQS

A3.7/5.0

Scored across 7 tools

Disambiguation4/5

Each tool targets a clearly different resource: Gulf jobs, Dubai property transactions, precious metals, Amazon rankings, Noon search, web search, and page conversion. The only mild overlap is between web_search (which can return page text when fetch_page_content is true) and web_to_markdown, which both can produce page content. amazon_best_sellers and noon_products are both commerce tools but differ in marketplace and function (rankings vs. keyword search), so they are distinguishable.

Naming Consistency4/5

All names are consistently lowercase snake_case with no camelCase or style mixing. However, the convention shifts between resource-noun names (gulf_jobs, gold_price, amazon_best_sellers, noon_products, dubai_property_transactions) and verb-style action names (web_search, web_to_markdown), so there is no single verb_noun pattern throughout.

Tool Count5/5

Seven tools is well within the ideal 3-15 range and each tool corresponds to a distinct data source or utility with no filler. The set is tightly scoped around Gulf/regional data plus general web grounding utilities.

Completeness4/5

For a read-only data aggregation server, the surface covers jobs, property transactions, metals, two marketplaces, and generic web retrieval, which is solid lifecycle coverage for its stated purpose. Minor gaps exist (e.g., property listings rather than only registered transactions, no non-gold FX or broader regional datasets), but web_search and web_to_markdown let an agent work around most of them.

Maintenance

ActivityMaintained
ResponsivenessNo issues