Skip to main content

Lore

Lore is an autonomous knowledge layer for engineering teams. It captures the unwritten rules, tribal knowledge, and operational context that live in Slack threads, PR discussions, and hallway conversations — then synthesizes them into structured, searchable documentation that stays in sync with your codebase.

How it works​

  1. Ingest — Lore connects to where your team already communicates. Mention @lore in a Slack thread to capture a conversation, or connect your GitHub repos to automatically ingest PR discussions. You can also push text directly from the CLI.
  2. Synthesize — Raw conversations pass through an LLM pipeline that extracts structured documents. Duplicate detection and smart ingestion ensure your knowledge base stays clean.
  3. Surface — Engineers query Lore through the interface that fits their workflow: the dashboard, the CLI, the REST API, or an MCP server that gives AI agents direct access to your team's context.

Choose your path​

Slack
Slack IntegrationInstall the bot, mention @lore in threads to capture conversations.
GitHub
GitHub IntegrationConnect repositories to automatically ingest PR discussions and reviews.
CLI
CLI ReferenceAuthenticate, capture text, and manage workspaces from your terminal.
API
API ReferenceQuery documents and search your knowledge base programmatically.
MCP
MCP IntegrationConnect AI agents to Lore via the Model Context Protocol.

Quick start​

Install the CLI and authenticate:

npm install -g @lore/cli
lore login

This opens your browser to authenticate and select a workspace. Then capture your first piece of knowledge:

lore capture "Our staging deploys run on the :30 cron. Prod is manual via Slack."

For the full walkthrough, see the Quickstart Guide.