# DeepSift Search

> DeepSift Search is a freemium MCP server for web search and scraping. Web search for agents that returns only the answering passages, in Tavily's format. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Freemium
- **Platforms:** API, CLI
- **Website:** https://apify.com/aadstra/deepsift-search
- **Repository:** https://github.com/pavanmulka/deepsift
- **MCP Registry name:** io.github.pavanmulka/deepsift-search
- **Categories:** web search and scraping
- **Listing updated:** 2026-11-14

## About

Rather than returning whole pages, the hosted actor searches Google and Yahoo at once, opens the best results and uses a small AI model to keep the passages that best answer the question. Results are merged so that if one engine is slow or down, the other still answers, and filters cover country, language, date range and website.

It speaks the same paths and fields as the Tavily API (search, extract, crawl and map), so existing code can switch by changing the base URL and using an Apify token as the key. As an MCP server at aadstra--deepsift-search.apify.actor/mcp it exposes web_search, web_read, web_map and web_crawl. web_read returns the text of up to 20 pages or PDFs, optionally only the parts about a topic, and web_crawl reads many pages of one site keeping the relevant ones.

The same author's GitHub project is a local variant for Claude Code that uses your Claude subscription, and also adds a web_research tool.

## Features

- **Passage extraction:** Returns the best passages per page rather than full pages.
- **Google and Yahoo merged:** If one engine fails, results from the other still arrive.
- **Tavily-compatible API:** Same /search, /extract, /crawl and /map paths and response shape.
- **Search filters:** Country, language, date range and website filters.
- **Page reading:** Reads up to 20 pages or PDFs, optionally focused on a query.
- **Site crawl and map:** Crawl one site's pages or list its addresses.
- **Batch runs:** Paste a list of queries as a normal Apify run and get a table.

## Good for

- Replace a Tavily call in an existing RAG app
- Give Claude or Cursor live web search
- Read several pricing pages and keep only the relevant parts
- List every page on a documentation site
- Run a batch of search queries into a table

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http deepsift-search https://aadstra--deepsift-search.apify.actor/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://aadstra--deepsift-search.apify.actor/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "deepsift-search": {
      "url": "https://aadstra--deepsift-search.apify.actor/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "deepsift-search": {
      "type": "http",
      "url": "https://aadstra--deepsift-search.apify.actor/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "deepsift-search": {
      "serverUrl": "https://aadstra--deepsift-search.apify.actor/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "deepsift-search": {
      "type": "http",
      "url": "https://aadstra--deepsift-search.apify.actor/mcp"
    }
  }
}
```

## Tools

- `web_search`: Searches Google and Yahoo and returns the passages that best answer the question.
- `web_read`: Returns the text of up to 20 pages or PDFs, optionally only the parts about a topic.
- `web_map`: Lists a website's page addresses.
- `web_crawl`: Reads many pages of one site and keeps the relevant ones.

## FAQ

### How do I switch from Tavily?

Point the Tavily client's base URL at https://aadstra--deepsift-search.apify.actor and use your Apify token as the API key.

### Which MCP tools does it add?

web_search, web_read, web_map and web_crawl.

### Do I need a separate API key?

No. Your Apify token is the only key.

### Is there a self-hosted version?

The GitHub project runs locally for Claude Code and uses your Claude subscription, while the Apify version never does.

### Is DeepSift Search free?

DeepSift Search has a free plan, with paid plans for more usage or features.

### How do I add DeepSift Search to Claude Code?

Run this command in your terminal: claude mcp add --transport http deepsift-search https://aadstra--deepsift-search.apify.actor/mcp

### What are the best alternatives to DeepSift Search?

The closest alternatives to DeepSift Search are Scrapling MCP Server, Firecrawl MCP Server, SearXNG Search, NeuralVerge and AI-Powered directory.

## Alternatives

- [Scrapling MCP Server](https://ai-powered.com/mcp/scrapling-mcp-server): Scrapling MCP Server is an open-source MCP server for web search and scraping.
- [Firecrawl MCP Server](https://ai-powered.com/mcp/firecrawl-mcp-server): Firecrawl MCP Server is a freemium MCP server for web search and scraping.
- [SearXNG Search](https://ai-powered.com/mcp/searxng-search): SearXNG Search is an open-source MCP server for web search and scraping.
- [NeuralVerge](https://ai-powered.com/mcp/neuralverge): NeuralVerge is an MCP server for web search and scraping.
- [AI-Powered directory](https://ai-powered.com/mcp/ai-powered-directory): AI-Powered directory is a free MCP server for developer tools.
- [Shopify MCP Server (any store)](https://ai-powered.com/mcp/shopify-mcp-server-any-store): Shopify MCP Server (any store) is a paid MCP server for e-commerce.
- [Chanroll: YouTube Channel Video Scraper (Official API)](https://ai-powered.com/mcp/chanroll-youtube-channel-video-scraper-official-api): Chanroll: YouTube Channel Video Scraper (Official API) is a paid MCP server for design and media.
- [Vidvett: YouTube Video Details Scraper (Official API)](https://ai-powered.com/mcp/vidvett-youtube-video-details-scraper-official-api): Vidvett: YouTube Video Details Scraper (Official API) is a paid MCP server for design and media.

---
Source: https://ai-powered.com/mcp/deepsift-search
