🧠
BetaFrameworkHigh complexity

Pillar (Archon OS)

Knowledge + task management backbone for AI coding assistants — runs as an MCP server.

Model Context Protocol
Protocol
Claude Code, Kiro, Cursor, Windsurf
Compatible IDEs
Crawled web + uploaded docs/PDFs
Knowledge sources
Coolify-ready, SSL automatic
Deployment

What is Pillar (Archon OS)?

Pillar is the production fork of Archon, evolved into a knowledge + task management backbone for AI coding assistants. For a developer it is a sleek interface to manage knowledge, context, and tasks across projects. For the AI coding assistant, it is a Model Context Protocol (MCP) server that exposes that same knowledge so the AI can reason from grounded context instead of guessing.

You connect Claude Code, Kiro, Cursor, Windsurf — anything that speaks MCP — and the assistant gains access to your crawled documentation, uploaded PDFs, project tasks, and an advanced RAG layer that picks the right source for each query.

The vision is to replace the messy "paste your codebase into chat every time" workflow with a persistent, queryable, multi-project knowledge surface that any AI agent can ground itself in.

Who is it for?

  • Developers actively using AI coding assistants who want them to stop hallucinating
  • Teams running multiple projects who need a shared context layer
  • Anyone tired of pasting documentation into LLM chats over and over
  • AI agent builders who need a knowledge backbone they can plug into

What does it do?

  • Crawl any website and turn it into a queryable knowledge source
  • Upload PDFs / docs and they become MCP-discoverable instantly
  • Advanced RAG strategies for smart source selection per query
  • Task management integrated with the knowledge base
  • MCP server exposes everything to compatible AI IDEs
  • Real-time updates as you add content
  • One-click Coolify deployment with SSL + auto-routing

How is it built?

TypeScriptPythonSupabaseMCP serverarchon-mcp (port 8051)archon-agents (port 8052)archon-agent-work-orders (port 8053)DockerCoolify

What makes it interesting?

  • Persistent context across coding sessions — no re-pasting
  • Multi-IDE compatible via MCP (works with whichever AI tool you prefer)
  • Production-grade deployment template, not just a local toy
  • Knowledge + tasks unified — your AI sees both at once
  • Replaces the "Archon as agent-builder" v1 vision with a sharper context-layer focus

Want something like this for your business?

We adapt the patterns above to your stack on every engagement. The 15-minute discovery call is free — you leave with a plan regardless.