Firecrawl for Automated Web-to-Report Workflows

Use Stated's Firecrawl-backed scraping workflow to fetch known public webpages, structure their content into report blocks and refresh source-backed data over time.

2026-09-15 · 2 min read

Stated uses Firecrawl for workflows that fetch content from known public webpages and turn that source material into Stated Page content.

When to use it

Use this workflow when you already know the URL containing the information you need: a public pricing page, product page, documentation page, announcement or other public web source.

For broad research questions that require discovery across multiple sources, use Stated's grounded research workflow instead.

Fetch a public page

The scraping workflow accepts a public HTTP(S) URL. In the default single-page mode, Stated fetches that page and can run its structuring step to turn the source into appropriate text, table and chart blocks.

Charts are created only when the source contains real comparable numbers.

Source attribution

Blocks created from scraped content retain a visible source link and fetch timestamp. This provenance should remain attached when the content is used in a human-facing report.

Preview mode

Use preview mode when your automation needs the structured result without creating or modifying a Stated Page. The fetch still consumes the applicable provider credits.

Append to an existing Page

Pass an existing Page id, slug or Stated URL when the fetched content belongs in a persistent artifact. Appending to an existing Page does not consume another publication allowance.

Refresh source-backed blocks

Blocks originally created through the scraping workflow can be refreshed from their recorded source. Stated re-fetches the source and updates the block in place, preserving the Page and its existing interactions.

This is useful for changing public figures such as pricing, availability or other source-backed information.

Crawl limits

Use site crawling only when multiple pages are actually required. Keep max_pages conservative because credits are charged per fetched page.

Costs

The standard scraping workflow consumes Stated Credits for the Firecrawl fetch and, when enabled, an additional credit for Stated's structuring model. Refreshing an existing scraped block also consumes credits because it performs a real new fetch.

Check plan usage before bulk operations and avoid uncontrolled refresh loops.

Example automation

Firecrawl is the retrieval layer. Stated is the persistent publication and human-facing layer.