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

# HTTP Sources

> Connect HTTP sources to DataHarbor and handle authentication material safely.

# HTTP Sources

HTTP sources are one of the available source models in DataHarbor. In practice, this means REST-style endpoints that return JSON today, plus other payload formats like CSV, YAML, and Markdown that DataHarbor can normalize before controls run.

## What you provide

When you connect an HTTP source, you typically define:

* a base URL for the upstream service
* a request path template that describes the resource hierarchy
* the upstream authentication method, when one is required

See [Enrolling Data Sources](../core/enrolling-data-sources) for the current enrollment flow and path-template behavior.

## Authentication

The current docs model supports these upstream authentication patterns:

| Method     | Typical use                              |
| ---------- | ---------------------------------------- |
| Public     | Source does not require credentials      |
| API key    | Source expects a key in a request header |
| Basic auth | Source expects a username and password   |

## Credential storage

When you provide upstream credentials for an HTTP source, DataHarbor stores that secret material in a **secure secret store**.

That includes values such as:

* upstream API keys
* basic auth usernames and passwords
* other source-level secrets required to call the upstream service

In docs and examples, continue to use placeholders such as `YOUR_API_KEY`. Real credential values should not appear in MDX, checked-in config, or shared examples.

## How HTTP sources fit the pipeline

Once DataHarbor fetches an HTTP response, the rest of the request flow is the same:

1. normalize the payload into the canonical JSON model
2. match the correct object definition
3. apply controls and transforms
4. format the governed result for delivery

See [Data Pipeline](../core/data-pipeline) for the full flow and [Input Normalization](./input-normalization) for the format-specific parsing stage.

## Next steps

<CardGroup cols={2}>
  <Card title="Route Matching" icon="route" href="./route-matching">
    How DataHarbor validates requests against your path template
  </Card>

  <Card title="Input Normalization" icon="arrow-down-up" href="./input-normalization">
    See how HTTP payloads normalize before controls run
  </Card>

  <Card title="Enrolling Data Sources" icon="plug" href="../core/enrolling-data-sources">
    Review the current enrollment workflow
  </Card>
</CardGroup>