Why Truto
Truto is a Unified-API platform with four opinions that put it on the opposite side of the line from most of the category.
Pass-through, not a cache (by default)
By default, every Unified API call is a live call to the vendor. Truto doesn't run a background sync to mirror your customer's data into our database, doesn't return data from a Truto-side copy, and doesn't pretend to know more about the vendor than the vendor does. The trade-off is honest: you share the vendor's rate limit, and your latency tracks the vendor's. In return you don't reconcile a stale mirror, and there's no Truto-shaped data residency surface to argue about with your customer.
When you do want the cached experience that Merge, Finch, and other sync-first providers ship by default, the same Unified APIs are also available against a synced store — pick whichever side of the cache you want to own:
- RapidBridge — schedule a sync job that pulls Unified API resources on your cadence and streams every record to your webhook, your datastore (S3, GCS, MongoDB, Qdrant), or a SuperQuery table. Use this when the data needs to land in your warehouse or your database.
- SuperQuery — read from a Truto-managed synced store by hitting the same
/unified/...endpoints with?truto_super_query=apac|wnam. Populated by RapidBridge sync jobs with asuperquerydestination. Use this when you want the Merge/Finch-style "we serve from cache" shape without standing up your own pipeline.
See Realtime data model.
Declarative integrations and Unified APIs
New integrations and Unified APIs are added by editing config — auth, base URL, resource list, mapping rules — not by writing platform code. The same surface customers use to add their own integration or override a Unified API mapping is the surface Truto uses to ship the catalog. Learn why this is important.
See Creating integrations and Creating Unified APIs.
Admin-API complete
Everything in app.truto.one is also a route. Installing an integration, minting a link token, listing integrated accounts, rotating an API token, registering a webhook, scheduling a sync — all of it is an HTTP call. Anything you can do as a human in the Dashboard you can also do from a script, a Workflow, or a coding agent.
See API reference.
Coding-agent native
The platform ships with a skills pack and a CLI alongside the docs. An agent that supports the Agent Skills convention can wire a link-token route, a webhook handler, and the first Unified API call into your codebase from a single prompt — using the same routes documented here, not a sandboxed code generator.