Commonplace
The theory of LLM wikis, running as one.
Commonplace is a framework for agent-operated knowledge bases — LLM wikis in the sense Karpathy sketched: a persistent, linked markdown layer that AI agents build and maintain around your own work. It ships the type system, writing conventions, agent skills, and Python commands to run one.
A wiki is two things — notes and the links between them — and an LLM wiki is one where the agent produces both: it concretizes a vague thought into a committed note, then connects it to everything you've already written. That turns an ephemeral chat, where the insight scrolls away, into a durable, growing body of your thinking. The agent takes the two slow parts, drafting and filing; judging whether a claim is true still falls to you — though review gates and refinement loops are moving more of that into the agents too.
It is self-hosting, in the bootstrapping sense: the theory of how to build LLM wikis lives here as notes, and the agents that maintain this repository run on the methodology those notes describe. Nothing here is documentation about a separate system — the wiki is the system, and reading these pages is watching it run.
The content is AI-generated through human-AI collaboration: a human directs the inquiry, and AI agents (Claude, ChatGPT, and others) draft, connect, and maintain the notes.
The central theory is deploy-time learning: deployed AI systems improve through structured knowledge that accumulates in the repo alongside the code. Agents operate under bounded context — they can't load everything, so the knowledge they load must cover many situations. The notes here develop that theory and apply it to the design of agent-operated knowledge bases. See the Reference collection for architecture, types, and CLI.
Threads worth following
How agents learn. Deploy-time learning fills the gap between training and in-context learning. It starts with accumulation — capturing observations, decisions, and patterns as durable artifacts. Three operations transform what's accumulated: constraining narrows interpretation (conventions → structured types → deterministic code), distillation extracts focused artifacts from larger reasoning, and discovery produces the high-reach theories that are accumulation's most valuable items. The learning theory index maps the full landscape.
What makes memory agent-usable. Agent memory needs discoverable, composable, trusted knowledge under bounded context. Discoverable: agents find what they need without loading everything. Composable: notes chain into arguments via explicit link semantics. Trustworthy: notes declare their maturity (seedling → current) and link to the sources or notes they rest on, so readers know how much weight a claim holds.
Information and bounded observers. Information value is observer-relative — the same data can contain extractable structure for one observer and noise for another. This grounds why distillation creates value and why reverse-compression — expanding text without adding extractable structure — is a real failure mode.
Agent memory systems compared. We reviewed 129 agent memory systems — Mem0, Graphiti, Cognee, Letta, and more — with a comparative analysis across six architectural dimensions. The reviews were mostly agent-produced. The key finding — that the fundamental split is who decides what to remember, not storage format — emerged from an agent traversing linked reviews and spotting patterns across them.
Browse
Each collection has a README that serves as its curated landing — all are linked from the top menu. The current collections are Notes, Reference, Agent Memory Systems, Agentic Systems, Sources, Instructions, and Workshops.
For deeper navigation:
Use it yourself
Commonplace is open source. You can use it in two ways:
Clone and explore. The repo is a functioning knowledge base out of the box. Add notes alongside the existing ones, run the agent skills (/cp-skill-connect, /cp-skill-validate, /cp-skill-ingest), and build on the theory. This is also the right mode for evaluating the system before installing it elsewhere.
Install into your own project as a theory builder for your domain. Commonplace can be added to any project as a submodule or cloned subdirectory. It provides the type system, writing conventions, agent skills, and methodology — your agents accumulate domain knowledge and build explanatory structure instead of starting every session cold. See the installation guide for setup instructions and the installation architecture for design rationale.
Both paths use the same framework. The difference is whether you're building theory about agentic systems (this repo's domain) or about your own.
For the full technical reference — directory layout, prerequisites, scripts — see the README.