Skip to content

Skill

Scrape

gstack

Pulls structured data out of a web page that does not offer it cleanly - reading the page and extracting the fields you need. It gathers data from the web, unrelated to authoring documentation, maintaining it after a release, or rendering a document to PDF.

Situations it fits

You need the information from a web page that offers no download or API. You want it extracted off the page into a usable format.
Scrape. The two documentation skills are the trap. scrape extracts structured data from a web page that does not hand it over cleanly. By contrast document-generate writes documentation that does not exist yet, from scratch, for undocumented code; document-release updates EXISTING docs after a change ships, bringing them back in sync with what landed; make-pdf converts a Markdown file into a polished PDF - a format change, not a content change.
A competitor's site lists prices across dozens of pages with no API. You want those prices and product names pulled off the pages into structured data.
Scrape. The two documentation skills are the trap. scrape extracts structured data from a web page that does not hand it over cleanly. By contrast document-generate writes documentation that does not exist yet, from scratch, for undocumented code; document-release updates EXISTING docs after a change ships, bringing them back in sync with what landed; make-pdf converts a Markdown file into a polished PDF - a format change, not a content change.

3 scenarios in the bank answer to Scrape. The rest are in the game.

Skills it gets confused with

These share a family with Scrape, which is another way of saying they are the ones you might reach for by mistake.

Knowing what Scrape does is the easy half. Telling it apart from the others under time pressure is the game.

Today's session

Scrape is part of garrytan/gstack. Licence: MIT. The description above was written for this game, not taken from the skill.