Skip to content

io.github.AshutoshRaj97/agentready-mcp

by AshutoshRaj97 · remote · local (npm) · no sign-in to list tools · v1.3.1 · published by GitHub user AshutoshRaj97

Query any docs site via MCP. Submit a URL, ask questions, get cited answers.

Changes · 4
CompareRegistry entry

8 tools: 4 change things and 4 only read.

Scanned Oct 21, 2026

No check matched anything in the tool definitions. No finding is not a safety guarantee. How the checks work.

claude mcp add --transport http agentready https://www.agentready.it.com/api/mcp

Tools

Open a tool for its reasons and full description. Descriptions are the publisher's own words, shown as they are.

8 tools

Changes things 4Claude Code: ask
  • plan_site_actionTurn a natural-language request into a grounded, read-only AgentReady plan for a website.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Turn a natural-language request into a grounded, read-only AgentReady plan for a website. If the site is not indexed yet, AgentReady indexes it automatically first. This prototype never executes side effects; it identifies supported steps, sources, risks, and whether a future execution would require confirmation.
  • rate_answerRate the quality of a previous ask_site answer from 1 (not useful) to 5 (excellent).No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Rate the quality of a previous ask_site answer from 1 (not useful) to 5 (excellent). Pass the request_id returned in structuredContent when available so the rating can be tied to the exact answer.
  • refresh_siteForce a full re-crawl of a site to pick up new or changed content.Reads text others writeNo 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

    • untrusted-inputThe description's first sentence says "crawl": it reads web pages, which anyone can write.
    Description, as the server wrote it
    Force a full re-crawl of a site to pick up new or changed content. If the site has never been indexed, AgentReady performs the initial indexing automatically. Use when ask_site returns outdated information or when you know the site has recently been updated. Takes ~60 seconds. After completion, ask_site returns fresh content.
  • submit_siteIndex any website so it can be queried with ask_site.Can send data outNo findings

    Why it is in this group

    • The name "submit_site" says "submit": a word for a change.

    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
    Index any website so it can be queried with ask_site. Use this when the site is not yet in list_sites. Handles JS-rendered pages (React, Next.js, Vue SPAs) that web_fetch cannot read — uses a four-layer pipeline: llms.txt → HTTP+cheerio → __NEXT_DATA__ extraction → Jina Reader headless browser. Takes ~60 seconds. Once indexed, ask_site queries are instant.
Only reads 4Claude Code: allow, ask for 1 outbound
  • ask_siteQuery any website's documentation and get cited, multi-page answers in natural language.Reads text others writeCan send data outNo findings

    Why it is in this group

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

    What it can reach

    • untrusted-inputThe description's first sentence says "query" 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
    Query any website's documentation and get cited, multi-page answers in natural language. Use ask_site when you need: (1) answers that synthesize information across multiple pages of a site, (2) documentation from JS-rendered sites (React, Next.js, Vue SPAs) where web_fetch returns empty or partial HTML, (3) citations linking back to the exact source pages, (4) faster results than fetching and reading individual pages one by one. For sites not yet indexed, ask_site auto-crawls and answers in ~60s — no separate submit_site call needed.
  • get_site_capabilitiesReturn the AgentReady capability manifest for a website.Reads text others writeNo findings

    Why it is in this group

    • The name "get_site_capabilities" says "get": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "return" and "website": it reads web pages, which anyone can write.
    Description, as the server wrote it
    Return the AgentReady capability manifest for a website. If the site is not indexed yet, AgentReady indexes it automatically before returning its freshness, read-only limits, schemas, and available MCP and HTTP endpoints.
  • get_site_factsReturn structured facts extracted from a website — pricing, limits, features, contact info, and more.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_site_facts" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "return" and "contact": it reads contacts.
    • untrusted-inputThe description's first sentence says "return" and "website": it reads web pages, which anyone can write.
    Description, as the server wrote it
    Return structured facts extracted from a website — pricing, limits, features, contact info, and more. Facts marked as "owner-verified" have been explicitly attested by the site owner and are guaranteed accurate ground truth; use these in preference to scraped RAG answers for critical comparisons. Ideal for price comparison across multiple sites, feature matrix lookups, or any task where you need structured data rather than a natural-language synthesis.
  • list_sitesList all websites currently indexed in AgentReady.Reads text others writeNo findings

    Why it is in this group

    • The name "list_sites" says "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.
    Description, as the server wrote it
    List all websites currently indexed in AgentReady. Use this to check if a domain is already available before submitting it — indexed sites return instant cited answers via ask_site. The index covers developer tools, APIs, cloud platforms, frameworks, databases, and more, and grows as new sites are submitted.

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.

10 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 write5
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out2
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__agentready__get_site_facts"
    ]
  }
}

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.

Looked at Oct 22, 2026, 14:50 UTC, unchanged since Oct 22, 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
4
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
none
RFC 9728 protected resource metadata
none
Authorization servers
none
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 4 tools that change things (4 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://www.agentready.it.com/api/mcp: tools/list returned 8 tools to an anonymous client, including plan_site_action (write), submit_site (write), refresh_site (write), rate_answer (write)

    What this check looks for

Install

Pinned to version 1.3.1. Values in angle brackets are yours to fill in.

claude mcp add --transport http agentready https://www.agentready.it.com/api/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": [
      "mcp__agentready__list_sites",
      "mcp__agentready__get_site_capabilities",
      "mcp__agentready__get_site_facts"
    ],
    "ask": [
      "mcp__agentready__plan_site_action",
      "mcp__agentready__submit_site",
      "mcp__agentready__ask_site",
      "mcp__agentready__refresh_site",
      "mcp__agentready__rate_answer"
    ],
    "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 agentready. Review it before you use it.

Capability tags behind 6 rules
  • mcp__agentready__list_sites untrusted-input
  • mcp__agentready__get_site_capabilities untrusted-input
  • mcp__agentready__submit_site outbound
  • mcp__agentready__ask_site untrusted-input, outbound
  • mcp__agentready__refresh_site untrusted-input
  • mcp__agentready__get_site_facts private-data, untrusted-input