# Beever Atlas

A platform that turns a team's Slack, Discord, Teams and Mattermost conversations into a self-maintaining wiki, extracting facts and building a cited knowledge graph queryable in natural language or through MCP.

Beever Atlas pulls conversation history from Slack, Discord, Microsoft Teams and Mattermost, extracts atomic facts, deduplicates them, and clusters them into topic pages with citations back to the source messages. A graph store links the people, decisions and projects mentioned across channels, feeding both an auto-maintained wiki and a question-answering agent that streams cited answers.

Under the hood, a backend, a chat bot and a frontend are backed by Weaviate, Neo4j, MongoDB and Redis, with a six-stage ingestion pipeline distilling raw messages into a three-tier semantic store and a graph store. An MCP server exposes 28 tools with per-agent authentication so external assistants such as Claude Code and Cursor can query the same knowledge base.

Website: https://docs.beever.ai/atlas