agentready-mcp
This server connects AI agents to AgentReady, a hosted service that indexes websites and makes them queryable with natural language. It provides three core tools:
submit_site: Submit any public website URL for indexing. Supports static sites, server-rendered pages, and JavaScript SPAs, with content available for querying in ~60 seconds.list_sites: Retrieve all websites currently indexed by AgentReady, including their domain, title, summary, and page count.ask_site: Ask a natural language question about a specific indexed website and receive a cited answer grounded in that site's content. If the site isn't already indexed, it will be crawled and indexed automatically before answering.
Plugin (via Starlight) that automatically indexes the documentation site in AgentReady on every build, ensuring fresh content for AI agents.
Plugin that automatically indexes the documentation site in AgentReady on every build, keeping the AI discovery layer up to date.
Extension that automatically indexes the documentation site in AgentReady on every build, maintaining an up-to-date AI-accessible index.
This server has a remotely hosted endpoint. Open its connector listing for connection details, authentication requirements, and setup instructions:
@agentreadyweb/mcp
What is AgentReady?
AgentReady is a hosted capability layer that makes any website discoverable and usable by agents through MCP.
Paste a URL → AgentReady crawls the site, generates a spec-compliant llms.txt, and hosts a live /ask RAG endpoint and MCP server. Any MCP-compatible agent client can then discover the site, ask natural-language questions, inspect capabilities, and create grounded read-only plans with citations.
The problem it solves: AI agents using web_fetch fetch one page at a time, get empty HTML from JavaScript SPAs (React, Next.js, Vue), and hallucinate when the answer spans multiple pages. AgentReady indexes the whole site, handles JS rendering, and retrieves across pages — so agents get the right answer instead of a confident wrong one.
What's already indexed: Browse the live AgentReady directory → or index any public site yourself.
Key properties:
Works on any public URL — static sites, React/Next.js SPAs, Docusaurus, GitBook, custom engines
No account required to index your first site
Shared index — one team member submits a site, everyone on the team can query it instantly
Handles JS-rendered pages that
web_fetchreturns empty forCapability manifests expose freshness, schemas, endpoints, and read-only limits
Grounded plans return evidence, risks, confirmation requirements, and durable receipts
Related MCP server: sofya-mcp
CLI
The same package doubles as a CLI — no install, no account:
# Agent-readiness report card (llms.txt, sitemap, robots, JS-rendering, index status)
npx @agentreadyweb/mcp grade yourdocs.com
# Ask any site a question, get a cited answer (auto-indexes new sites in ~60s)
npx @agentreadyweb/mcp ask stripe.com "what is the test card number?"
# Index or re-crawl a site
npx @agentreadyweb/mcp index yourdocs.com
npx @agentreadyweb/mcp refresh yourdocs.comFor CI, dashboards, or scripts, add --json to receive the raw report on
stdout (progress remains on stderr):
npx @agentreadyweb/mcp grade yourdocs.com --json | jq '.grade, .score'The command still exits 1 when the grade is below B.
grade exits non-zero below a B, so you can use it as a CI gate. refresh in your docs deploy pipeline keeps the index fresh automatically:
# GitHub Actions — after your docs deploy step
- run: npx @agentreadyweb/mcp refresh yourdocs.comConnect any MCP-compatible client to AgentReady:
Local MCP bridge (stdio)
For clients that use an mcpServers configuration, add the AgentReady bridge:
{
"mcpServers": {
"agentready": {
"command": "npx",
"args": ["-y", "@agentreadyweb/mcp"]
}
}
}Restart your client. You'll have eight tools available:
list_sites— see all indexed websitesget_site_capabilities— index on demand, then inspect a site manifest, freshness, schemas, and endpointsget_site_facts— key structured facts extracted from a siteask_site— query any site with cited, multi-page answersplan_site_action— index on demand, then create a grounded, read-only plan and receiptsubmit_site— index any website so it can be queriedrefresh_site— re-crawl a site, or perform its initial index when it is newrate_answer— submit quality feedback
Cursor
Add to ~/.cursor/mcp.json:
{
"agentready": {
"command": "npx",
"args": ["-y", "@agentreadyweb/mcp"]
}
}Project or terminal setup
npx @agentreadyweb/mcpOr add the same agentready server to your client's project configuration to share it with your team:
{
"mcpServers": {
"agentready": {
"command": "npx",
"args": ["-y", "@agentreadyweb/mcp"]
}
}
}VS Code (GitHub Copilot agent mode)
Requires VS Code 1.99+ with the GitHub Copilot extension. Create .vscode/mcp.json in your project root:
{
"servers": {
"agentready": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@agentreadyweb/mcp"]
}
}
}Copilot's MCP tools are only available in agent mode. Commit this file to share with your team.
Windsurf / Zed / other clients
Any MCP client that supports stdio transport works the same way — use npx -y @agentreadyweb/mcp as the command.
WebMCP (no install)
If your client supports HTTP transport, connect directly by URL — no npm required:
https://www.agentready.it.com/api/mcpDocusaurus plugin
If your docs site uses Docusaurus, auto-index on every build:
npm install @agentreadyweb/docusaurus-plugin// docusaurus.config.js
plugins: [
['@agentreadyweb/docusaurus-plugin', { domain: 'docs.yoursite.com' }]
]MkDocs plugin
If your docs site uses MkDocs, auto-index on every build:
pip install mkdocs-agentready# mkdocs.yml
plugins:
- search
- agentreadyIf site_url is not set, specify the domain explicitly:
plugins:
- agentready:
domain: docs.yoursite.comStarlight plugin
If your docs site uses Starlight (Astro), auto-index on every build:
npm install starlight-agentready// astro.config.mjs
import agentready from 'starlight-agentready'
export default defineConfig({
site: 'https://docs.yoursite.com',
integrations: [
starlight({
plugins: [agentready()],
}),
],
})Sphinx extension
If your docs site uses Sphinx, auto-index on every build:
pip install sphinx-agentready# conf.py
extensions = [
"sphinx_agentready.extension",
]
# Domain is inferred from html_baseurl automatically, or set explicitly:
agentready_domain = "docs.yoursite.com"Available tools
list_sites
Lists all websites currently indexed by AgentReady with their titles and page counts. Use this to check if a domain is already available before submitting it.
get_site_capabilities
get_site_capabilities(domain: string)Returns the site's capability manifest, freshness state, schemas, and available HTTP/MCP endpoints.
get_site_facts
get_site_facts(domain: string)Returns key structured facts extracted from a site — pricing tiers, feature lists, limits, and other high-signal data points that agents frequently need.
plan_site_action
plan_site_action(domain: string, request: string)Creates a grounded, read-only plan with evidence, risks, confirmation requirements, and a durable receipt. It does not execute side effects.
submit_site
submit_site(url: string)Index any website with AgentReady. Takes ~60 seconds. Handles static sites, server-rendered pages, and JavaScript-heavy SPAs via a four-layer pipeline. Once done, query it with ask_site.
Example: submit_site("https://docs.example.com")
ask_site
ask_site(domain: string, query: string, url?: string)Ask a question about any website and get a cited answer grounded in its content. Synthesises information across multiple pages. If the site isn't indexed yet, AgentReady crawls and indexes it automatically before answering (~60s).
Example: ask_site("stripe.com", "What are the fees for card payments?")
refresh_site
refresh_site(domain: string)Force a full re-crawl of a site to pick up new or changed content. If the site is new, AgentReady performs its initial index automatically. Takes ~60 seconds.
Example: refresh_site("docs.example.com")
rate_answer
rate_answer(domain: string, rating: number, request_id?: string, comment?: string)Submits 1–5 quality feedback, optionally tied to the exact ask_site request.
Deploy webhook
Automatically re-index your docs on every deploy. No auth required — rate limited to once per hour per domain.
curl -X POST https://www.agentready.it.com/api/webhook/refresh \
-H "Content-Type: application/json" \
-d '{"domain": "docs.yoursite.com"}'Or pass the domain as a query param (works with Vercel/Netlify form-encoded webhook payloads):
https://www.agentready.it.com/api/webhook/refresh?domain=docs.yoursite.comHow indexing works
AgentReady uses a layered approach to handle any public website:
llms.txt / llms-full.txt — if the site publishes one, it's used as a high-quality structured content source
Standard HTML crawl — fetches up to 10 pages via sitemap or link crawling, extracts clean text
__NEXT_DATA__extraction — for Next.js apps, parses server-side rendered data embedded in the HTMLJina Reader fallback — for JS-only SPAs that return empty HTML, uses a remote rendering service to extract content
Sites behind authentication or with no public HTML content cannot be indexed.
Environment variables
Variable | Default | Description |
|
| Override the MCP endpoint (for self-hosted) |
|
| Set to |
Telemetry
This package sends a lightweight anonymous ping to agentready.it.com/api/telemetry each time the MCP bridge starts or a CLI command runs. It contains: a random install ID (stored in ~/.agentready/install-id), package version, Node version, platform, and which mode/command was used. No query content, no domain names, no personal data. Set AGENTREADY_TELEMETRY=0 to opt out.
Available Tools
2 toolsask_siteAInspect
Ask a question about any website indexed by AgentReady and get a cited answer grounded in that site's actual content. First call list_sites to see which domains are available.
Inputs: query (required), domain (required).
list_sitesAInspect
List all websites indexed by AgentReady that can be queried with ask_site. Returns domain, title, summary, and page count for each site.
Inputs: no named parameters.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 2 tools
ask_site and list_sites have clearly distinct purposes: one lists available sites, the other queries a specific site's content. There is no overlap.
Both tool names follow a consistent verb_noun pattern (list_sites, ask_site) and use snake_case.
With only 2 tools, the server is minimal but covers the core workflow (discovery and query). It feels slightly thin but acceptable for a narrow domain.
The tool surface covers the primary tasks: discovering indexed sites and asking questions about them. Minor missing features like pagination for list_sites or site detail, but no dead ends.
Maintenance
Related MCP Connectors
Web data for agents: YouTube transcripts, screenshots, Google News, WHOIS, jobs, tech stack, more.
Web data for agents. Extract web pages, crawl websites, discover URLs, and search the web.
Public web tools for agents: product extraction, claim checks, webpage QA and ranked audits.
Live web checks for AI agents: sitemaps, robots.txt, URL status, broken links, feeds, citations.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceProvides AI agents and assistants with advanced web crawling and RAG capabilities, enabling them to scrape websites and perform semantic search over crawled content.1MIT
- AlicenseAqualityBmaintenanceWeb tools for AI agents. Search the web for full page content, fetch URLs as clean markdown including PDFs, extract structured data from a page with a prompt, and run multi-source deep research that returns a cited report.42MIT
- AlicenseAqualityBmaintenanceProvides AI agents live web access through scraping, multi-engine search, site mapping, crawling, SEO audits, and 31 data collectors via natural language tool calls.964 npmMIT
- AlicenseAqualityDmaintenanceEnables AI agents to extract structured content from URLs, search the web via DuckDuckGo, monitor site availability, and convert HTML to markdown without requiring API keys.552 npmMIT