Beever Atlas
Turn team and agent chat into a typed knowledge graph and auto wiki so your coding agent retrieves consistent product memory from MCP.
Overview
Beever Atlas is a MCP server for the Build phase that converts team chat into a typed LLM knowledge graph and auto-generated wiki for agent retrieval over stdio Docker.
What is this MCP server?
- Open-source LLM knowledge base from team chat into structured graph data
- Auto-generated wiki surfaces for human-readable docs
- stdio MCP via Docker OCI ghcr.io/beever-ai/beever-atlas:0.2.0
- Python module entry: beever_atlas.api.mcp_server
- Version 0.2.0 with GitHub source at Beever-AI/beever-atlas
- Published OCI version: 0.2.0
- Transport: stdio via Docker python -m beever_atlas.api.mcp_server
- MCP manifest schema: 2025-12-11
Community signal: 378 GitHub stars.
What problem does it solve?
Builders lose decisions buried in chat logs while agents hallucinate specs because there is no structured, queryable team memory.
Who is it for?
Small teams and solo builders who want a local or Docker-hosted team brain linked to Claude Code, Cursor, or Windsurf.
Skip if: One-off landing copy with no ongoing chat corpus, or teams that only need a static Notion export with no MCP agent loop.
What do I get? / Deliverables
After you run the Atlas MCP container and connect your agent, conversations become graph-backed wiki knowledge your agent can cite during implementation.
- Running stdio MCP server from the Beever Atlas container
- Typed knowledge graph derived from ingested conversations
- Auto-generated wiki pages agents can query during builds
Recommended MCP Servers
Journey fit
Knowledge capture and living documentation peak while you build—before launch content and long-run ops wikis diverge from what the team actually said. Beever Atlas is an LLM knowledge base with graph + wiki output—canonical shelf is Build → docs, not a one-off research scrape.
How it compares
Self-hosted knowledge-graph MCP, not a payment rail or static code linter.
Common Questions / FAQ
Who is Beever Atlas for?
Indie developers and lean teams using AI coding agents who need chat and decisions converted into a searchable graph and wiki via MCP.
When should I use Beever Atlas?
Use it during Build docs when you are consolidating product knowledge from chats before agents implement features or you ship external documentation.
How do I add Beever Atlas to my agent?
Pull ghcr.io/beever-ai/beever-atlas:0.2.0, configure stdio MCP with python -m beever_atlas.api.mcp_server per the registry packageArguments, then register the server in Claude Code, Cursor, or your MCP client.