Zeno Wiki — A Plug-and-Play Second Brain for AI

A persistent, compounding knowledge base that travels with you — plug it into Claude, a voice AI agent, a Discord bot, or any project that needs a brain.

---

What Is This?

Zeno Wiki is a file-based second brain built on the LLM Wiki Pattern. Instead of dumping documents into a RAG pipeline and re-deriving answers every time, Zeno compiles knowledge once into a structured, interlinked wiki of markdown files — then exposes that wiki as an MCP server that any AI can call.

The result: knowledge that compounds. Every new source makes every existing answer richer. Cross-references are pre-built. Synthesis already happened. You ask a question and the answer draws on everything you’ve ever learned — not just the top-k chunks that matched your query.

This repo contains two things:

1. The Obsidian vault — the knowledge base itself (Zeno - 100x's Second Brain/)
2. The MCP server — a Cloudflare Worker that exposes the wiki as tools any AI can use (zeno-mcp-server/)

---

The Problem with RAG

Most AI knowledge pipelines work like this:

Question → Embed query → Find top-k chunks → Stuff into context → Answer

Every question starts from zero. The LLM re-discovers the same facts from raw documents on every call. Ask something that needs synthesis across five sources and it has to piece together fragments on the fly — and probably misses some.

Zeno does something different.

New source → LLM reads + synthesizes → Updates wiki pages → Knowledge persists

Question → LLM reads index → Reads relevant pages → Answers from compiled knowledge

Knowledge is compiled once. Contradictions are flagged when they’re discovered. Cross-references are wired up. The wiki grows smarter with every source you add — not just bigger.

---

The OPT Framework — Why This Exists

Zeno was built using the OPT Framework (Operating Model → Processes → Tasks) — a method for systematically identifying what to automate with AI before you build anything.

The operating model for Zeno: build a knowledge base that contains everything — cohort notes, company philosophy, product decisions, frameworks — and make it available to any AI project without rebuilding it each time.