Authenticate the source
Use scoped credentials, signature verification, tenant context, and explicit authorization for each integration path.
This page describes the exchange model we are building. There is no public endpoint to call today, and no published schema, credential type, or delivery commitment. If automated intake is on your roadmap, tell us what you need to send and when.
ReturnValue treats every request and event as a governed exchange with an authenticated source, defined schema, idempotent handling, and reviewable delivery history.
Use scoped credentials, signature verification, tenant context, and explicit authorization for each integration path.
Reject or quarantine unsupported versions, missing fields, invalid types, duplicates, and out-of-order events.
Expose receipt, processing, retry, dead-letter, and resolution status without leaking sensitive payload details.
A typed API for pushing data into ReturnValue, and webhooks for reacting to changes inside it, are in active development with design partners. Neither surface is available yet; this page reflects what is being designed, not what is shipped. There is no public endpoint to call today, and we are not shipping a generic API kit until it holds up against real use.
The ingest API is designed as a typed endpoint for pushing structured records or files into a workspace, with the same schema discipline as a manual upload. Outbound webhooks are designed for subscribing to changes inside a workspace, such as an insight regenerated, a dataset updated, or an audit event recorded.
A generic create-read-update-delete API is easy. A useful one holds up against the patterns teams actually run: nightly batch ingest, change-driven webhooks into a downstream system, signed payloads that support the receiver security review. We are designing against those patterns rather than a feature checklist. This page will say available when the surface clears that bar, not before.
A failed delivery becomes a visible operational exception rather than a silent hole in the evidence timeline.
An integration is dependable when success and failure are equally visible.
Illustrative view. Values shown are examples, not customer data.