Ben Miller benmillerat

LLM Wiki v2

A pattern for building personal knowledge bases using LLMs. Extended with lessons from building agentmemory 10K Stars ⭐️, a persistent memory engine for AI coding agents.

This builds on Andrej Karpathy's original LLM Wiki idea file. Everything in the original still applies. This document adds what we learned running the pattern in production: what breaks at scale, what's missing, and what separates a wiki that stays useful from one that rots.

What the original gets right

The core insight is correct: stop re-deriving, start compiling. RAG retrieves and forgets. A wiki accumulates and compounds. The three-layer architecture (raw sources, wiki, schema) works. The operations (ingest, query, lint) cover the basics. If you haven't read the original, start there.

LLM Wiki

A pattern for building personal knowledge bases using LLMs.

This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.

The core idea

Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.

	## Workflow Orchestration

	### 1. Plan Mode Default
	- Enter plan mode for ANY non-trivial task (3+ steps or architectural decisions)
	- If something goes sideways, STOP and re-plan immediately
	- Don't keep pushing.
	- Use plan mode for verification steps, not just building
	- Write detailed specs upfront to reduce ambiguity

	### 2. Subagent Strategy

	You are Manus, an AI agent created by the Manus team.

	You excel at the following tasks:
	1. Information gathering, fact-checking, and documentation
	2. Data processing, analysis, and visualization
	3. Writing multi-chapter articles and in-depth research reports
	4. Creating websites, applications, and tools
	5. Using programming to solve various problems beyond development
	6. Various tasks that can be accomplished using computers and the internet

	Having tried a couple of the previous shortcut solutions for iOS I was convinced something easier was possible using the API. The below is my first proper go at putting something together.

	This shortcut will be available when "sharing" a URL from within a browser on iOS. On sharing to this shortcut it will automatically pull all available tags from the Linkding server and allow multi-tagging for the URL.

	Limitations:
	- Currently only previously defined tags can be used
	- Title will rely on the automatic process on the server, manual override isn't available
	- The same applies to the description
	- Tags are currently listed in the order they were created. Alphabetical would make more sense.

	// Lizenz: AGES Dashboard COVID19 AT
	// 3-stellige GKZ verwenden (zweite Spalte in https://covid19-dashboard.ages.at/data/CovidFaelle_GKZ.csv).
	// Widget Parameter: "204,KFL;312;706" für Klagenfurt Land (angezeigt als KFL), Korneuburg und Landeck
	//
	// Basiert auf der deutschen Variante von
	// - kevinkub https://gist.github.com/kevinkub/46caebfebc7e26be63403a7f0587f664
	// - Baumchen https://gist.github.com/Baumchen/6d91df0a4c76c45b15576db0632e4329
	//
	// Angaben ohne Gewähr.