Skip to content

Cornerstone · Start here

What Cornerstone is for

Cornerstone is the contextual knowledge layer for your AI.

Your organisation's rules, decisions, processes and policies are spread across documents, folders and people's heads. An AI that cannot see them gives generic answers. An AI handed all of them at once loses the detail and costs more to run. Cornerstone sits between the two: it holds your documents, turns them into short facts, and gives your AI only the few facts a question needs, each with the document it came from.

What you get

  • Answers with their sources. Every answer names the document each fact came from, so you can open it and check.
  • It tells you when it does not know. If nothing recorded answers the question well enough, it says so and shows what it searched, instead of guessing.
  • Current facts win. When a decision or rule is replaced, the old one is kept on record but no longer used in answers.
  • People only see what they are allowed to. A fact is shown only to people who can read every document it came from.

Who it is for

Anyone whose work depends on knowing how the organisation works: what was decided, what the rule is, which version is current. You use it in the Cornerstone web app, or through the AI you already use, such as Claude, ChatGPT, Gemini, Copilot or a model you run yourself.

What it is not for

  • It does not write answers in its own words. There is no AI model inside Cornerstone; it returns your facts and their sources, and your AI does the writing.
  • It is not a file store for everything. Load the documents people ask about, not the whole archive.
  • It is not where code-level detail belongs. Your code repositories stay the place for that.

Updated 14 November 2026