What the Knowledge Engine is
A governed store of facts about your product, your competitors and your market, each one tied to where it came from.
The Knowledge Engine is what stops a draft describing a feature you do not ship.
It holds facts, each attached to an entity and traced to a source, with a status saying how much it can be trusted. Drafting runs read from it rather than from whatever a language model happens to remember about your category.
The three parts
- Entityown_product | competitor | category
The subject a fact is about: your own product, a competitor, or the category you both sit in.
- Source
Where a fact came from. Either a web page anyone can open, or a document you uploaded privately.
- Fact
One claim, with a facet (
pricing,features,integrations,limits,positioningorguidance), a status, and the source behind it.
Why the source kind matters
This is the mechanism that lets you upload an internal one pager without it turning into a footnote on your blog.
What it is used for
- Grounding drafts, so specifics come from your product rather than from a plausible guess.
- Comparisons, so competitor claims are traceable rather than invented.
- Keeping published posts honest over time, by noticing when a fact behind an article has gone stale.