Developer docs

Plug AI agents into verified data

Everything you need to turn client domains into live knowledge nodes, budget fuel, and connect agents over MCP. No API keys required on the read side.

Core concept

Cyclone Engine

A Cyclone is one sync of a client domain. The engine crawls the site, distills a structured summary and target keywords, regenerates the domain's llms.txt feed, and reports two separate scores:

Sync Health · Infrastructure

Confirms llms.txt, JSON-LD and IndexNow are live and that the six ingestion engines (Exa, Tavily, Perplexity, Brave, Firecrawl, IndexNow) accepted the node.

Brand Mention Share · Organic

Queries the domain's target keywords on Perplexity, Exa, Tavily and Brave and measures how many answers actually cite the brand or domain. Probes are free.

If one engine times out or errors, it is logged as failed and the rest of the cycle completes normally.

Cyclones run automatically on each domain's schedule (checked every hour) or on demand with the Trigger Cyclone button in the console. Every result is streamed into the live console log.

Get started

Agency Quickstart

  1. 1Create an account. You start with 5 free fuel credits.
  2. 2Open the console and add a client domain. Daily sync is the default: about 30 credits a month per domain.
  3. 3Trigger the first Cyclone. The public node page and llms.txt feed go live immediately.
  4. 4Share the white-label report with your client from the domain row.
  5. 5Top up fuel before credits run out so scheduled syncs never pause.
Deploy the Cyclone

Billing

Fuel Economics

Each Cyclone burns exactly 1 fuel credit. When a balance hits zero, scheduled syncs pause until you top up. Nodes stay public in the meantime.

FrequencySyncsPer domain / mo
Daily1 / day~30 credits
Every 6 hours4 / day~120 credits
Hourly24 / day~720 credits

Eco

$19

30 credits

Turbo

$49

120 credits

Agency

$199

500 credits

Discovery

llms.txt & Sitemap

Every active node serves plain text at a stable URL, refreshed on each Cyclone. All live nodes are listed in the sitemap so ExaSearchBot, PerplexityBot and other AI crawlers find them automatically.

GET https://vortexnode.dev/node/{slug}/llms.txt
Content-Type: text/plain; charset=utf-8

GET https://vortexnode.dev/sitemap.xml
GET https://vortexnode.dev/robots.txt

Own-domain authority

Connect Client Root Domain (/llms.txt)

Serve the live VortexNode feed directly at clientdomain.com/llms.txt so AI crawlers attribute it to the client's own root domain. Replace {slug} with the node slug, or open any domain in the console for snippets with the slug already filled in.

serveshttps://clientdomain.com/llms.txtfromhttps://vortexnode.dev/node/{slug}/llms.txt
worker.js
export default {
  async fetch(request) {
    const url = new URL(request.url);
    if (url.pathname !== '/llms.txt') return fetch(request);

    const upstream = await fetch('https://vortexnode.dev/node/{slug}/llms.txt', {
      cf: { cacheTtl: 3600, cacheEverything: true },
    });

    return new Response(upstream.body, {
      status: upstream.status,
      headers: {
        'Content-Type': 'text/plain; charset=utf-8',
        'Cache-Control': 'public, max-age=3600',
        'X-Robots-Tag': 'all',
      },
    });
  },
};

Deploy as a Worker, then add the route clientdomain.com/llms.txt to it. Zero maintenance: the edge caches the feed for one hour.

Agents

MCP Server (JSON-RPC)

A public JSON-RPC 2.0 endpoint exposing the search_verified_nodes tool. It supports initialize, tools/list and tools/call. Add it to any MCP-capable client:

{
  "mcpServers": {
    "vortexnode": { "url": "https://kuasvlixegganngfrcxp.supabase.co/functions/v1/mcp" }
  }
}

Or call it directly:

curl -X POST https://kuasvlixegganngfrcxp.supabase.co/functions/v1/mcp \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "tools/call",
    "params": {
      "name": "search_verified_nodes",
      "arguments": { "query": "logistics", "limit": 3 }
    }
  }'

Try it live

Live MCP Probe · search_verified_nodes
live

Request · POST /functions/v1/mcp

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "search_verified_nodes",
    "arguments": {
      "query": "",
      "limit": 3
    }
  }
}

Response

// send a request to see the live response

Lead capture

Embed Widget

Drop the Instant AI Visibility Audit onto any site. Visitors scan their own domain and every scan is recorded as a lead.

<iframe
  src="https://vortexnode.dev/embed"
  style="width:100%;min-height:420px;border:0;border-radius:8px;"
  loading="lazy"
  title="AI Visibility Audit"
></iframe>

A ready-to-copy snippet is also available in console settings. Preview the widget at /embed.

Browse every verified node in the public directory.

Open directory