LLM wiki
Also known as: agent knowledge base
A knowledge base written for a language model: plain markdown entity pages linked with wiki-links that an agent builds and maintains, so knowledge compounds. Lighter than RAG.
An LLM wiki, a term popularized by Andrej Karpathy, is a set of markdown entity pages written for a model rather than a person, linked with wiki-links so they form a connected graph. An agent reads it directly and folds in new sources over time, so what it learns compounds instead of evaporating between sessions.
This site is built in that shape: every entry is an entity page, every cross-reference is a wiki link, and each page is available as raw markdown for a model to read. For a trading agent, the same pattern can hold its strategies, markets, and backtest results.
Related
- Loop engineering Designing the control loop an agent runs in (gather context, decide, act, observe, repeat) rather than hand-writing each prompt. A trading agent is a high-stakes example.
- TradingAgents Popular open-source multi-agent LLM framework that simulates a trading firm (analysts, researchers, trader, risk debate). Research and backtest focused, not a live-execution layer.