Skip to main content
This guide walks you through the fastest path to agentic commerce readiness: audit your domain, generate your UCP manifest, publish it, and install the attribution snippet. By the end you’ll have a live .well-known/ucp file and AI-sourced sessions appearing in your analytics.

​
Before you start

You need:
  • Your store domain (e.g. yourstore.com)
  • An Asva AI API key — get one at asva-ai.com/get-help
  • Ability to host a static file at /.well-known/ucp on your domain
Set your API key as an environment variable before running any of the examples below: 
export ASVA_API_KEY="asva_live_..."
1

Run a Readiness Score

Before implementing anything, audit your domain to see exactly what’s missing. The Readiness Score returns a 0–100 score, a letter grade, and a prioritized list of gaps to close.
curl -X POST https://asva-ai.com/api/audit \
  -H "Authorization: Bearer $ASVA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"domain": "yourstore.com"}'
Example response:
{
  "domain": "yourstore.com",
  "score": 42,
  "grade": "C",
  "gaps": [
    { "id": "well_known_missing", "severity": "critical", "label": ".well-known/ucp not found" },
    { "id": "product_schema_incomplete", "severity": "high", "label": "Product JSON-LD schema missing on 6 pages" },
    { "id": "dark_traffic_untracked", "severity": "medium", "label": "Asva attribution snippet not detected" }
  ],
  "breakdown": {
    "discoverability": 60,
    "catalog": 45,
    "checkout": 30,
    "attribution": 0
  }
}
Review the gaps array — start with critical severity items first.
2

Generate your UCP manifest

Generate a compliant .well-known/ucp manifest for your domain. You need to provide the URLs for your catalog and checkout endpoints (you’ll build these out fully in the UCP guide).
curl -X POST https://asva-ai.com/api/manifest \
  -H "Authorization: Bearer $ASVA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "domain": "yourstore.com",
    "catalog_endpoint": "https://yourstore.com/api/ucp/products",
    "checkout_endpoint": "https://yourstore.com/api/ucp/checkout"
  }'
Example response (your manifest file):
{
  "version": "1.0",
  "domain": "yourstore.com",
  "capabilities": {
    "product_catalog": {
      "endpoint": "https://yourstore.com/api/ucp/products",
      "format": "json",
      "version": "1.0"
    },
    "checkout": {
      "endpoint": "https://yourstore.com/api/ucp/checkout",
      "supported_methods": ["POST"],
      "version": "1.0"
    }
  }
}
Save this JSON — you’ll publish it in the next step.
3

Publish the manifest

Host the manifest at /.well-known/ucp on your domain. This is how Google AI Mode, Gemini, and other agents discover your commerce capabilities.
# Static host / Vercel / Netlify
mkdir -p public/.well-known
curl -X POST https://asva-ai.com/api/manifest \
  -H "Authorization: Bearer $ASVA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"domain":"yourstore.com","catalog_endpoint":"...","checkout_endpoint":"..."}' \
  -o public/.well-known/ucp
Then verify the file is publicly accessible:
curl https://yourstore.com/.well-known/ucp
# Should return your manifest JSON
Serve this file with Content-Type: application/json and no authentication — it must be publicly readable. Any auth requirement will prevent agents from discovering your capabilities.
4

Install the attribution snippet

Add this snippet to your storefront <head> to surface AI-sourced sessions as a distinct channel in GA4. Replace YOUR_PROPERTY_ID with the property ID from your Asva dashboard.
<!-- Asva AI Attribution -->
<script>
  (function(w,d,s,l,i){w[l]=w[l]||[];
  w[l].push({'asva.start': new Date().getTime(), event:'asva.js'});
  var f=d.getElementsByTagName(s)[0],
  j=d.createElement(s),dl=l!='asvaLayer'?'&l='+l:'';
  j.async=true;j.src='https://cdn.asva-ai.com/attribution.js?id='+i+dl;
  f.parentNode.insertBefore(j,f);
  })(window,document,'script','asvaLayer','YOUR_PROPERTY_ID');
</script>
The snippet analyzes session fingerprints, timing patterns, and UTM signals to identify AI-sourced visits that would otherwise appear as “Direct” in GA4. See Dark traffic attribution for how it works.
5

Re-run the score to confirm

Run the audit again after publishing your manifest and installing the snippet. A passing UCP manifest plus attribution snippet should bring your score above 60.
curl -X POST https://asva-ai.com/api/audit \
  -H "Authorization: Bearer $ASVA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"domain": "yourstore.com"}'
If well_known_missing is gone from gaps and attribution is no longer 0 in the breakdown, you’re on track. Continue with the UCP guide to build out your catalog and checkout endpoints.

​
Next steps

Implement UCP endpoints

Build the catalog and checkout endpoints that back your manifest.

Implement ACP for ChatGPT

Add ChatGPT Instant Checkout support via Agentic Commerce Protocol.

Set up attribution

See your AI-sourced traffic in GA4 and your own analytics.

API reference

Full API docs — auth, readiness, manifest, attribution.