All About Lore

Lore is knowledge capture for AI agents. It interviews employees, turns the unwritten rules of their work into sourced rules a manager approves, and serves those rules to your AI agents through an API.

You add the person in Lore and send them a private interview link. Lore chats with them in plain text for about 30 to 45 minutes and keeps asking about exceptions, approvals, clients and workarounds until the specifics come out. They can pause and come back any time.

When you add someone, choose whether they are leaving or staying. For people who are staying, the interview captures what they know so colleagues and agents can rely on it without always asking them.

Your agent sends a question to one endpoint, POST /api/v1/query, with an API key. It gets back an answer plus the approved rules behind it and who each rule came from. You can also export every approved rule as JSON or Markdown and load it into the tools you already run.

A wiki only holds what someone wrote down. Lore goes after the exceptions people never write down, and every rule keeps the condition, the action, the reason and the exact quote it came from, so your team can check it.