Workflows
Create, update, delete and query Workflows.
Last updated
Create, update, delete and query Workflows.
Last updated
When ingesting content into Graphlit, workflows are used to configure the processing of the content into the Graphlit Platform. This includes ingestion filtering, audio transcription, entity extraction, and web scraping (via link crawling).
With the CLI, you can configuring one or more workflow stages, and then reuse the workflow across content and feeds.
Here we are creating a workflow to crawl web links from ingested web pages.
Here we are creating a workflow to extract text from PDFs using Azure Document Intelligence.
Once you've created a workflow, you can update it in-place, and reconfigure as appropriate.
Here we're selecting 'allow content domain', so the web scraping will include links with the same domain as the original web page.