Stream Stripe charges to Datadog as business metrics

Build me a deterministic, code-based workflow that pumps Stripe charges into Datadog as live business metrics and dashboard events. No reasoning is needed once the field map is set.

Trigger: a Stripe webhook on the charge.succeeded and charge.failed events.

For each incoming event, parse the Stripe charge payload and extract: charge id, customer id, amount (in the smallest currency unit), currency, payment_method, payment_method_details.type, status (succeeded or failed), and failure_code plus failure_message when present.

Then do these two things in parallel:

1) Call Datadog Submit Metrics (V2) to send two time-series for this charge. First, a counter metric named stripe.charges.count with value 1, tagged with currency, payment_method, and outcome (succeeded or failed). Second, a gauge metric named stripe.charges.amount with value equal to the Stripe amount in the smallest currency unit, tagged with the same currency, payment_method, and outcome. Use the current event timestamp as the metric timestamp.

2) Call Datadog Post an Event with a one-line title describing the charge (for example, "Stripe charge succeeded: 4200 usd" or "Stripe charge failed: card_declined"), a body containing the charge id and customer id (and the failure_code and failure_message if the charge failed), alert_type set to info for succeeded charges and warning for failed charges, and the same currency, payment_method, and outcome tags. This makes each charge appear as an annotation on Datadog dashboards.

The point of the workflow is to let engineering correlate revenue regressions with infra issues on a single Datadog dashboard. Strict field map, no judgment calls in the workflow itself.

Integrations: Stripe (webhook trigger) and Datadog (Submit Metrics V2 and Post an Event).