# Agent Read Proxy

> Agent Read Proxy is a paid MCP server for web search and scraping. Pay-per-call page reader for agents: Markdown, tables, feeds and screenshots. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Paid
- **Platforms:** API
- **Website:** https://agent-read-proxy.agent-read-proxy.workers.dev
- **MCP Registry name:** dev.workers.agent-read-proxy.agent-read-proxy/agent-read-proxy
- **Categories:** web search and scraping
- **Listing updated:** 2026-11-15

## About

Payment is the unusual part: there are no accounts or API keys, and each call is paid in USDC on Base through the x402 protocol. A plain page read costs $0.002, a batch of up to five URLs $0.006, metadata, links and feeds $0.001, tables and sitemaps $0.002, and the JavaScript-rendering read and screenshots $0.005. Failed fetches are not charged.

The reader strips scripts, menus and sidebars and returns text or Markdown with title, author, date, language and canonical URL. It also handles PDFs, Word, Excel, OpenDocument and CSV files up to 10 MB. The rendered mode loads a page in a real browser first, which helps with shops and single-page apps.

Limits are public http and https pages on standard ports, 2 MB downloaded, a 10 second timeout and 50,000 characters of text; private and local addresses are refused. MCP clients need an x402-enabled client such as @x402/mcp.

## Features

- **Clean text or Markdown:** Returns readable content with title, author, date and language.
- **Document support:** Reads PDF, docx, xlsx, OpenDocument and CSV files up to 10 MB.
- **Batch reads:** Reads up to five URLs in parallel for $0.006.
- **Table extraction:** Returns every HTML table as JSON.
- **Rendered pages:** Loads JavaScript-built pages in a real browser first.
- **Screenshots:** Returns a JPEG or PNG, optionally full page up to 8,000 pixels.

## Good for

- Feed an agent clean page text for research
- Ingest web pages as Markdown for a RAG pipeline
- Pull prices or stats tables from a page as JSON
- Monitor a page or a feed for changes

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http agent-read-proxy https://agent-read-proxy.agent-read-proxy.workers.dev/mcp
```

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

```sh
https://agent-read-proxy.agent-read-proxy.workers.dev/mcp
```

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

```json
{
  "mcpServers": {
    "agent-read-proxy": {
      "url": "https://agent-read-proxy.agent-read-proxy.workers.dev/mcp"
    }
  }
}
```

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

```json
{
  "servers": {
    "agent-read-proxy": {
      "type": "http",
      "url": "https://agent-read-proxy.agent-read-proxy.workers.dev/mcp"
    }
  }
}
```

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

```json
{
  "mcpServers": {
    "agent-read-proxy": {
      "serverUrl": "https://agent-read-proxy.agent-read-proxy.workers.dev/mcp"
    }
  }
}
```

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

```json
{
  "mcpServers": {
    "agent-read-proxy": {
      "type": "http",
      "url": "https://agent-read-proxy.agent-read-proxy.workers.dev/mcp"
    }
  }
}
```

## Tools

- `read_page`: Reads one web page or document as text or Markdown.
- `read_pages`: Reads several pages in one call.
- `page_metadata`: Returns a page's metadata.
- `page_links`: Lists the links on a page.
- `page_tables`: Extracts HTML tables as JSON.
- `read_feed`: Gets the latest posts from an RSS or Atom feed.
- `site_sitemap`: Lists a site's page URLs from its sitemap.
- `read_rendered_page`: Reads a page after rendering it in a browser.
- `page_screenshot`: Takes a screenshot of a page.

## FAQ

### Do I need an API key?

No. Agents pay per call in USDC on Base via x402.

### What does a page read cost?

$0.002 per page; batches of five cost $0.006.

### Am I charged when a fetch fails?

No, failed calls are not charged.

### What are the size limits?

2 MB downloaded, 10 seconds, 50,000 characters of text.

### Is Agent Read Proxy free?

No. Agent Read Proxy is a paid product.

### How do I add Agent Read Proxy to Claude Code?

Run this command in your terminal: claude mcp add --transport http agent-read-proxy https://agent-read-proxy.agent-read-proxy.workers.dev/mcp

### What are the best alternatives to Agent Read Proxy?

The closest alternatives to Agent Read Proxy 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/agent-read-proxy
