# Kevin Liao — agents don't need memory, they need documentation

> Why a Markdown 'brain' of specs and decisions beats a vector database of chat snippets for coding agents.

- Org: Kevin Liao
- Date: 2026-10-03
- Category: article
- Source: https://liao.gg/blog/agents-dont-need-memory
- Canonical page: https://ai-tldr.dev/releases/kevin-liao-agents-dont-need-memory/

## In short

Kevin Liao argues that memory plugins for coding agents are just RAG over old chats, and that agents work better with a folder of plain Markdown docs they read before and update after each task. The essay reached 324 points on Hacker News.

## What it is

In 'Agents Don't Need Memory. They Need Documentation.', Kevin Liao says every agent memory plugin works the same way: it cuts session transcripts into snippets, stores them in a RAG database and injects the top five most similar snippets into each prompt. He calls the whole memory plugin ecosystem a solution to the wrong problem.

## How it works

Liao lists five problems with recall: snippets are picked by similarity, not correctness; they lose context and motivation; old conversations are treated as truth while the code keeps changing; an agent cannot search for what it does not know it is missing; and a store of thousands of embeddings cannot be audited. His alternative changes the loop from prompt → build → forget to prompt → consult → build → update, with the agent keeping instructions, specs, decisions, research and indexes in a structured folder.

## Why it matters

Liao's argument is that a single AGENTS.md file is not enough, and that plain documents can be read, edited, committed and shared by the team, unlike a hidden vector store. He has packaged the approach as Operator Memory, a free BSD-3-Clause plugin with no embeddings or background summarizers, which he says he has used for over a year.

## Who it's for

developers who use coding agents

## Links

- [The Essay](https://liao.gg/blog/agents-dont-need-memory)
- [HN Discussion](https://news.ycombinator.com/item?id=49945933)
- [Operator Memory](https://github.com/aerovato/operator-memory)
