> ## Documentation Index
> Fetch the complete documentation index at: https://docs.clinia.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Data sources

> Define the upstream systems that own records, schemas, and synchronization rules.

Data sources are the authoritative entry points for everything you ingest into the Clinia registry. Each source captures where a record originates, which profiles it must respect, and how your upstream system pushes changes into Clinia.

## What a data source governs

* **Provenance** — Every record, relationship, and object references its originating source so you can audit lineage and route updates back to the right system of record.
* **Synchronization flow** — The source declares how data arrives (currently API push) and which environments or jobs publish into it.
* **Processing boundary** — Validation, ingestion pipelines, and access controls execute independently per source, allowing teams to tune quality gates without affecting other domains.

<Warning>
  The Clinia platform currently only supports API-driven synchronization.
</Warning>

## Configuration essentials

A data source definition consists of a unique key, metadata, and the profiles it owns. Keep keys stable because they appear in API paths, identifiers, and audit logs.

```json theme={null}
{
  "key": "ehr-prod",
  "type": "registry",
}
```

* `key` — URL-safe identifier referenced across registry APIs and partitions. Must follow this pattern: `^[a-zA-Z0-9-][\w-]{0,61}$`
* `type` — Categorizes how data is managed. Use `registry` for resources written directly to the Clinia registry.

<Note>
  The Clinia platform only supports `registry` sources at this time.
</Note>

## Operating data sources

* **One source per schema** — When upstream systems expose different schemas or lifecycle rules, create dedicated sources. This keeps validation strict and provenance clear.
* **Coordinate schema changes** — Version profiles before pushing breaking changes so ingestion never fails unexpectedly.
* **Secure access** — Tie API keys and IAM roles to specific sources to prevent accidental cross-domain writes.
* **Monitor freshness** — Track ingestion timestamps to surface sources that stopped publishing data.

## Related references

* [List the data sources API](/api-reference/sources/list-the-data-sources) for the canonical schema and endpoints.
* [Data ingestion overview](/explanation/ingestion-pipeline/basics) to understand how sources feed the registry.

## Keep exploring

<CardGroup cols={2}>
  <Card title="Profiles" icon="identification" href="/explanation/sources/profile">
    Learn how schemas define which properties each source can store.
  </Card>

  <Card title="Relationships" icon="arrows-right-left" href="/explanation/sources/relationship">
    See how sources describe graph connections between their profiles.
  </Card>

  <Card title="Data model overview" icon="chart-network" href="/explanation/data-model/data-sources">
    Connect source configuration to the broader Clinia data model.
  </Card>
</CardGroup>