Jina - Read URL action
The Read URL action fetches a URL with Jina Reader and returns clean content in Jina.
SHARED READER OPTIONS
Most input fields on this action are also available in the Search web action, because both actions send the same Jina Reader headers.
Input
The following are a representative set of input fields.
This action has additional optional input fields inherited from Jina Reader's header options available in the recipe editor. Refer to the Jina Reader documentation for more information on available input fields.
| Input field | Description |
|---|---|
| URL | Enter the absolute URL to read. |
| Return format | Optional. Select the response format. Options are Markdown, HTML, Text, Screenshot, or Pageshot. The default is Markdown. |
| Engine | Optional. Select the Reader engine, such as Browser or Direct. |
| Timeout (seconds) | Optional. Enter a timeout limit, in seconds. Valid values are 1-180. |
| Target selector | Optional. Enter a CSS selector for the main content, for example article or .main. |
| No cache | Optional. Select Yes or No to bypass the cache and fetch fresh content. |
| Retain images | Optional. Select how images are kept in the output. Options are All, None, Alt, All_p, or Alt_p. |
| Retain links | Optional. Select how links are kept in the output. Options are All, None, Text, or GPT OSS. |
| Token budget | Optional. Enter a token budget for the response. |
| PDF page | Optional. Enter a PDF page index to read. |
| Custom output fields | Optional. Add fields to create datapills for output fields that the default response doesn't include. These fields aren't sent in the request. |
Output
The following are a representative set of output fields. Refer to the Jina Reader documentation for more information on available output fields.
| Output field | Description |
|---|---|
| Code | The response's HTTP-style status code. |
| Status | Jina's internal status code. |
| Data | The page's content. |
| Title (Data) | The page's title. |
| Description (Data) | The page's meta description. |
| URL (Data) | The page's URL. |
| Content (Data) | The page's extracted content, in the format selected by Return format. |
| HTTP status (Data) | The origin page's HTTP status code. |
| Meta | Token usage for the request. |
Last updated: