Skip to content
MCP server

Hydrafetch

Remote MCP server that turns any URL into clean Markdown or JSON for agents

Install
Pricing
Freemium
Platforms
API, Web
Runs
Remotely
Version
1.0.1
Listing updated
2 days ago

About Hydrafetch

Hydrafetch is a freemium MCP server for web search and scraping. It runs as a hosted remote server, so there is nothing to install locally.

Hydrafetch is built around a single response envelope. Each call returns the page content together with a confidence score and flags such as complete and blocked, so a pipeline can retry, review or drop a fetch before it reaches an index. The score is computed from the extraction itself, with no model in the loop.

The API covers scraping a single URL, mapping every URL on a site, crawling a whole site as an asynchronous job with a webhook, running a search whose results arrive already fetched and cleaned, and extracting typed records against a JSON schema. A separate brand endpoint takes a domain and returns logos, ranked colours, fonts and socials, each with its source and confidence.

The MCP server is hosted at api.hydrafetch.com/mcp over streamable HTTP, and the site says every endpoint is exposed as a tool. Official clients exist for six languages, and plain HTTP works too. Billing is per page: one credit for a delivered page, five for model-backed work such as schema extraction and summaries, and failed pages are never charged.

Install

Run in a terminal

claude mcp add --transport http web https://api.hydrafetch.com/mcp

Add to Settings → Connectors → Add custom connector

https://api.hydrafetch.com/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "web": {
      "url": "https://api.hydrafetch.com/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "web": {
      "type": "http",
      "url": "https://api.hydrafetch.com/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "web": {
      "serverUrl": "https://api.hydrafetch.com/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "web": {
      "type": "http",
      "url": "https://api.hydrafetch.com/mcp"
    }
  }
}

What it does

Self-scoring responses
Every response carries a confidence value and completeness and blocked flags.
One response shape
Scrape, map, crawl, search and extract all return the same envelope.
Asynchronous crawls
Site crawls run as a job with depth, path and subdomain controls and a webhook on completion.
Search with fetched results
Queries return ranked results already fetched and cleaned, with time, region and domain filters.
Schema extraction
Point a JSON schema at a page or site and get typed records with per-field confidence.
Brand data
A domain returns logos for light and dark, ranked colours, fonts and socials.
Pay per delivered page
Errors, blocks and dead ends cost nothing; JS rendering and retries add no extra charge.

Good for

  • Feed an agent clean Markdown from live web pages during a conversation
  • Crawl a documentation site into a retrieval index and drop low-confidence pages
  • Pull structured records, such as listings, from many pages into a fixed schema
  • Fetch a company's logo, colours and fonts to render in an app
  • Run a search and get cited, already-cleaned context in one round trip

Pricing

Freemium

500 free credits on signup; paid plans priced per 10K credits (Pilot: $8.17 per 10K, 60,000 credits a month)

Prices change. Check hydrafetch.com before you buy.

Hydrafetch FAQ

How is usage billed?

A successfully delivered page costs one credit. Schema extraction and summaries cost five, search costs one plus one per scraped result, and a styleguide costs ten. Failed pages are never billed.

Is there a free way to try it?

Signing up with a work email gives 500 free credits on the full API, no card required. They stay valid for 30 days.

What happens if I exceed my plan?

Usage continues at the plan's overage rate per 10,000 credits instead of stopping mid-job. The Pilot plan lists $11 per 10K in overage.

Which languages have official clients?

The site lists six: Node, Python, Go, Ruby, Rust and PHP. Plain HTTP with an X-API-Key header also works.

Does it handle JavaScript-rendered pages and bot walls?

The site says client-side apps come back as content and challenge pages are handled, and that a page which truly cannot be retrieved is reported as such.

Is Hydrafetch free?

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

How do I add Hydrafetch to Claude Code?

Run this command in your terminal: claude mcp add --transport http web https://api.hydrafetch.com/mcp

What are the best alternatives to Hydrafetch?

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

Alternatives to Hydrafetch

  1. MCP server for AI clients to scrape sites with HTTP and stealth browser sessions

    Runs locally
    • 17 tools
    • Open source
    • 104K GitHub stars
    • Web search and scraping
  2. MCP server that lets AI agents search, scrape and interact with web pages

    • 3 tools
    • Remote
    • Freemium
    • 7.8K GitHub stars
    • Web search and scraping
  3. SearXNG Search

    MCP server

    MCP server for web search and page reading through a SearXNG instance

    Runs locally
    • 3 tools
    • Open source
    • 1.4K GitHub stars
    • Web search and scraping
  4. NeuralVerge

    MCP server

    Structured, source-tagged business and people data from 300+ sources for agents

    • 200 tools
    • Remote
    • Free trial
    • 1 GitHub star
    • Web search and scraping
  5. Search AI tools, agents and MCP servers and get install configs from inside your AI assistant.

    • 4 tools
    • Remote
    • Free
    • Developer tools
  6. Hosted MCP server that returns any Shopify store's products, prices and stock

    Not checked yet
    • 4 tools
    • Remote
    • Paid
    • E-commerce