1 · Choose the production pattern
Assurance journey
Governance and reliability contract
Catchpoint runnable reference
Prove an AI commerce action
works—and stops safely.
An outside-in assurance reference for customer-facing agentic commerce. It executes deterministic Store Sync and Agent Ready journey contracts, verifies governance outcomes, and generates Catchpoint API and Playwright test blueprints.
No real funds
2 · Inspect execution evidence
Journey boundaries
Not run
Run a contract to inspect requests, responses, classifications, and assertions.
3 · PayPal alignment
What this does—and does not claim
Publicly aligned
Agent Ready, Store Sync, catalog/cart integration, Braintree or Orders payment boundary, and merchant order confirmation.
To validate with PayPal
Internal agent gateway, identity evidence, policy decision points, risk interfaces, audit fields, and ownership boundaries.
Catchpoint responsibility
Execute from the appropriate nodes, assert control and business outcomes, measure latency, and isolate the failing boundary.
Outside scope
Judging fraud-model accuracy, replacing governance engines, production payments, or exposing customer/payment data.
4 · Operationalize
Generate Catchpoint tests
Generate five JavaScript API canaries and one Playwright journey for the selected production pattern. Scripts use this deployed HTTPS origin automatically.
Create directly in Catchpoint
The Catchpoint REST key remains server-side. Enter the separate publisher key, validate access to product 1289415, then create the six tests as inactive. Their global nodes and frequency inherit from that product.