How AI agents capture knowledge through skills and memory, why humans still write some skills, and what self-improving means in code — based on the source of six open agent harnesses.

Self-improving AI agents sound like marketing — until you read the code. This is how agents capture knowledge as skills and memory, why a human still has to write some of it down, and what self-improving actually means once you strip away the hype: bookkeeping, an LLM review pass, and a human gate.

Skills and Memory: How AI Agents Actually Learn

A deep dive into the AI agent harness — context management, tool contracts, safety boundaries, persistence and recovery — based on reading the source code of six open agent harnesses.

We read the source code of six AI agent harnesses — Pi, Hermes, OpenCode, Kimi Code, OpenAI's Codex CLI and Google's Gemini CLI. This is what the harness really does for the model, why almost every agent failure is a harness failure, and why the engineering lives there, not in the model.

Tag: Context Engineering

Tag: Context Engineering

Skills and Memory: How AI Agents Actually Learn

What an AI Agent Harness Actually Does