Skip to content

Agent Read Proxy MCP server

dev.workers.agent-read-proxy.agent-read-proxy/agent-read-proxyagent-read-proxy.agent-read-proxy.workers.devstreamable-httpv1.5.0Scanned

Read any web page (even JS sites), PDFs, tables, RSS & screenshots for AI agents. Pay via x402.

9 tools: 2 change things and 7 only read.

Tools: 0 delete, 0 irreversible, 2 write, 7 read.
  • Reads private data · 1
  • Reads text others write · 7
  • Can send data out · 9
  • No findings

Tools listed without sign-in. Remote posture

With Anlyon

2 tools here change things. Govern them.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the Agent Read Proxy MCP server can do

The Agent Read Proxy MCP server is published by agent-read-proxy.agent-read-proxy.workers.dev, a domain the MCP Registry verified. It has 9 tools. It can create or change data (2 tools: page_links and page_tables), but cannot delete data. 7 tools only read.

1 tool reads private data (read_pages); 7 tools bring in text other people can write (read_page, read_pages, page_metadata and 4 more); 9 tools can send data out of your system (read_page, read_pages, page_metadata and 6 more). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.

None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.

Its remote endpoint lists its tools without sign-in, and its tools changed 2 times since Anlyon Radar first saw them.

Last scanned Nov 16, 2026, version 1.5.0.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

9 tools

Changes things 2Claude Code: ask
  • page_tablesExtract every HTML table on a web page as JSON: caption, header row and data rows as arrays of cell text.Can send data outNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Extract every HTML table on a web page as JSON: caption, header row and data rows as arrays of cell text. Good for prices, stats, schedules, rankings and spec sheets. Price: $0.002 USDC (x402).
Only reads 7Show 7 read-only tools
  • page_metadataGet a web page's metadata without its text: title, description, preview image, favicon, language, canonical URL, author and dates, Open Graph and Twitter card tags, RSS/Atom feed l…Reads text others writeCan send data outNo findings

    Why it is in this group

    • The description's first verb is "get": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "get" and "web": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Get a web page's metadata without its text: title, description, preview image, favicon, language, canonical URL, author and dates, Open Graph and Twitter card tags, RSS/Atom feed links, and its schema.org structured data (JSON-LD: products with prices and ratings, articles, recipes, events, organizations). Price: $0.001 USDC (x402).
  • page_screenshotTake a screenshot of a web page in a real browser and get the image back (JPEG by default, or PNG).Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "page_screenshot" says "screenshot": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "web": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Take a screenshot of a web page in a real browser and get the image back (JPEG by default, or PNG). Choose the window size, or full_page=true for the whole page (up to 8,000 pixels tall). Price: $0.005 USDC (x402).
  • read_feedGet the latest posts from a blog or news site as JSON (title, link, date, author, summary).Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "read_feed" says "read": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "get" and "post": it reads posts, which other people can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Get the latest posts from a blog or news site as JSON (title, link, date, author, summary). Give it a feed URL, or just the site's page and it finds the RSS/Atom feed automatically. Price: $0.001 USDC (x402).
  • read_pageFetch a public web page and return its readable content as JSON: plain text or Markdown (headings, lists, links kept), plus title, description, author, publish date, site name, lan…Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "read_page" says "read": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "fetch" and "web": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Fetch a public web page and return its readable content as JSON: plain text or Markdown (headings, lists, links kept), plus title, description, author, publish date, site name, language and preview image. Ads, menus and scripts are stripped. Also reads PDF, Word, Excel and CSV links. Failed fetches are not charged. Price: $0.002 USDC (x402).
  • read_pagesRead up to 5 public web pages or documents (PDF, Word, Excel, CSV) in one call, in parallel.Reads private dataReads text others writeCan send data outNo findings

    Why it is in this group

    • The name "read_pages" says "read": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "read" and "document": it reads documents.
    • untrusted-inputThe description's first sentence says "read" and "web": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.urls), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.urls): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Read up to 5 public web pages or documents (PDF, Word, Excel, CSV) in one call, in parallel. Returns each one's readable text or Markdown plus metadata, and lists any that failed. Charged only if at least one URL was read. Price: $0.006 USDC (x402).
  • read_rendered_pageLike /v1/read, but loads the page in a real browser first, so content built by JavaScript (online shops, web apps, single-page sites) is included.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "read_rendered_page" says "read": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "browser": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Like /v1/read, but loads the page in a real browser first, so content built by JavaScript (online shops, web apps, single-page sites) is included. Use it when /v1/read comes back empty or incomplete. Returns the readable text or Markdown plus metadata. Price: $0.005 USDC (x402).
  • site_sitemapList a website's page URLs (with last-modified dates) from its sitemap.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The description's first verb is "list": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "list" and "website": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    List a website's page URLs (with last-modified dates) from its sitemap. Give it any page on the site: it checks robots.txt and /sitemap.xml, follows sitemap indexes, and reads .xml.gz sitemaps. Price: $0.002 USDC (x402).

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

63 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 1 private-data tool breaks every one.

Reads private data1
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write7
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out9
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
deny to break every path
{
  "permissions": {
    "deny": [
      "mcp__agent-read-proxy__read_pages"
    ]
  }
}

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 15, 2026, 13:21 UTC, unchanged since Nov 8, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
2
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • WarnThe whole server

    The endpoint served tools/list without credentials, and the list has 2 tools that change things (2 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://agent-read-proxy.agent-read-proxy.workers.dev/mcp: tools/list returned 9 tools to an anonymous client, including page_links (write), page_tables (write)

    What this check looks for

Posture changes

  1. Nov 8, 2026
    • Tools that change things, listed without credentials: 0 to 2

Findings

Matches of fixed checks, with the text they matched.

No check matched anything in these tool definitions. That is not a safety guarantee.

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. PostureRemote posture changed. Tools that change things, listed without credentials: 0 -> 2.View diff
    • Tools that change things, listed without credentials: 0 to 2
  2. Tools2 added.+2 addedView diff
  3. Tools5 added.+5 addedView diff

Install

Pinned to version 1.5.0. Values in angle brackets are yours to fill in.

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

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [],
    "ask": [
      "mcp__agent-read-proxy__read_page",
      "mcp__agent-read-proxy__read_pages",
      "mcp__agent-read-proxy__page_metadata",
      "mcp__agent-read-proxy__page_links",
      "mcp__agent-read-proxy__page_tables",
      "mcp__agent-read-proxy__read_feed",
      "mcp__agent-read-proxy__site_sitemap",
      "mcp__agent-read-proxy__read_rendered_page",
      "mcp__agent-read-proxy__page_screenshot"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as agent-read-proxy. Review it before you use it.

Capability tags behind 9 rules
  • mcp__agent-read-proxy__read_page untrusted-input, outbound
  • mcp__agent-read-proxy__read_pages private-data, untrusted-input, outbound
  • mcp__agent-read-proxy__page_metadata untrusted-input, outbound
  • mcp__agent-read-proxy__page_links outbound
  • mcp__agent-read-proxy__page_tables outbound
  • mcp__agent-read-proxy__read_feed untrusted-input, outbound
  • mcp__agent-read-proxy__site_sitemap untrusted-input, outbound
  • mcp__agent-read-proxy__read_rendered_page untrusted-input, outbound
  • mcp__agent-read-proxy__page_screenshot untrusted-input, outbound

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/dev.workers.agent-read-proxy.agent-read-proxy/agent-read-proxy/verify?hash=29e687b91aee02ccb20911d89e8f7cf97550667bea6eaba30f144358ba470606" | jq -e '.changed == false'

Pinned hash 29e687b91aee, from Nov 16, 2026, 03:17 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/dev.workers.agent-read-proxy.agent-read-proxy/agent-read-proxy/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for dev.workers.agent-read-proxy.agent-read-proxy/agent-read-proxy
[![Anlyon Radar](https://radar.anlyon.com/badge/dev.workers.agent-read-proxy.agent-read-proxy/agent-read-proxy.svg)](https://radar.anlyon.com/servers/dev.workers.agent-read-proxy.agent-read-proxy/agent-read-proxy)

Source: MCP Registry. Synced Nov 2, 2026, 19:17 UTC.

Agent Read Proxy MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the Agent Read Proxy MCP server safe?

No scan can promise that. What its definitions show: 9 tools, 2 of which can change, send or delete something, and 0 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the Agent Read Proxy MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the Agent Read Proxy MCP server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the Agent Read Proxy MCP server have?

9 tools: 2 write, 7 read. page_links, page_tables, read_page, read_pages, page_metadata, read_feed, site_sitemap, read_rendered_page and 1 more.

Does the Agent Read Proxy MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the Agent Read Proxy MCP server send data to outside services?

9 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): read_page, read_pages, page_metadata, page_links, page_tables, read_feed and 3 more.

Has the Agent Read Proxy MCP server changed its tools?

Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 2, 2026. Every change is on this page and in its Atom feed.