Loading…
Loading…
Reading billing state makes almost every support and sales agent better. Writing to billing is where you want a hard stop. FlozentAi treats that split as configuration: publish the read actions, mark refunds and subscription changes as high risk, and let a human approve each one against the exact amount shown.
| Definition | Import the Stripe OpenAPI specification and review the operations you need — it is large, so narrow it during review. |
|---|---|
| Credential | Restricted API key. Create one scoped to the resources you publish, and store it in your vault. |
| Stored as | A secret:// reference to your own vault. FlozentAi never holds the value. |
Action names are yours to choose during review — these are the shapes most teams end up with. Anything that writes is confirmation-gated unless you deliberately publish it otherwise.
| Action | Kind | What it does |
|---|---|---|
customer.get | Read | Resolve a customer by id or email. |
invoice.list | Read | List invoices and their payment status. |
subscription.get | Read | Read the current plan and renewal date. |
refund.create | Write | Issue a refund. High risk — approval shows the amount before execution. |
There is no field in the console that accepts a token or a password. You store the secret in your vault and give FlozentAi a reference; the runtime resolves it inside the execution layer, after the policy decision.
No. The model sees the action’s schema and the arguments. The credential is resolved from your vault inside the execution layer, after the policy decision, and is never part of the prompt or the trace.
Publish the refund action with confirmation required and review the amount on each approval. FlozentAi enforces that a human approved this specific proposal; it does not replace your own Stripe-side restricted key limits, which you should also set.
Draft support replies from live system data and keep a human on the send button. Approval-gated AI for support teams.
Read more ProductShip an AI assistant inside your own SaaS product, scoped per customer, without building the governance layer yourself.
Read more Customer facingAnswer order, stock and delivery questions from live commerce data with an AI agent scoped to each customer.
Read moreImport the specification, review what it found, and publish only what you want reachable.