Tasks
Fetch Webpage Contents
The Fetch Webpage Contents task fetches the contents of one or more webpages by URL, returning structured data including page metadata, text, highlights, and optional AI-generated summaries.
Use cases
- Fetch the content of a webpage for use in an AI prompt.
- Extract text from one or more URLs to save into a stub or pass to a GPT task.
- Retrieve webpage content with AI-generated summaries for quick analysis.
- Scrape and process webpage content for downstream tasks.
Basic usage
Parameters
urls required Array.<string>
An array of one or more URLs to fetch content from. URLs without a protocol prefix will automatically have https:// prepended.
result_options optional object
Controls what content is returned for each URL.
Show child attributes
full_text optional boolean
When true, the full extracted text content of each page is included in the result under the text field.
Default: false
highlights optional boolean
When true, the result includes a highlights array containing the most relevant excerpts from each page, along with a highlightScores array indicating the relevance score of each highlight.
Default: false
ai_summary optional object
Configuration for AI-generated summaries. When enabled, a summary is produced for each page and included under the summary field.
Show child attributes
enabled required boolean
Whether to enable AI summary generation.
prompt required string
The prompt used to instruct the AI when generating the summary. Required when enabled is true.
Result
Properties
contents Array.<Object>
An array of result objects, one per URL. Each object contains metadata about the page and optionally its content, highlights, and AI summary depending on which result_options are enabled.
contents[].id string
The unique identifier for the result, typically the URL of the page.
contents[].title string
The title of the webpage.
contents[].url string
The URL of the webpage.
contents[].author string | null
The author of the page content, if available. Returns null when no author is detected.
contents[].image string | null
The URL of the featured or Open Graph image for the page, if available.
contents[].favicon string | null
The URL of the favicon for the page, if available.
contents[].text string
The full extracted text content of the webpage. Only present when result_options.full_text is true.
contents[].highlights Array.<string>
An array of the most relevant text excerpts from the page. Only present when result_options.highlights is true.
contents[].highlightScores Array.<number>
An array of relevance scores corresponding to each entry in highlights. Only present when result_options.highlights is true.
contents[].summary string
An AI-generated summary of the page content. Only present when result_options.ai_summary.enabled is true.
Examples
Fetch with full text and highlights
Retrieve the full page text and relevant highlights.
Result:
Fetch with AI summary
Enable AI summaries to get a concise description of each page.
Result: