Reading Hub

← Back to articles

#agents

共 7 篇。

  • Harness Engineering for Coding Agent Users

    10 min read 📘 10 语块

    A practical framework for building the outer harness around coding agents — feedforward guides that steer before the agent acts, and feedback sensors that help it self-correct.

  • Agent Harness Engineering

    13 min read 📘 10 语块

    A coding agent is the model plus everything you build around it. Harness engineering treats that scaffolding as a real artifact — and it tightens every time the agent slips.

  • Harness Engineering for Coding Agent Users

    8 min read 📘 10 语块

    A practical framework for building the outer harness around coding agents — feedforward guides that steer before the agent acts, and feedback sensors that help it self-correct.

  • Memory and Dreaming for Self-Learning Agents

    10 min read 📘 12 语块

    Anthropic PM Mahes walks through memory as a file system for agents, optimistic concurrency for multi-agent teams, and Dreaming — a new out-of-band process that synthesizes learnings across sessions to make tomorrow's agents smarter than today's.

  • Agent Harness vs Everything Else: The Real Difference

    7 min read 📘 12 语块

    A clear breakdown of what an agent harness actually is, how it differs from frameworks like LangChain, and the nine components every modern harness needs.

  • How to Make Claude Code Your AI Engineering Team — Garry Tan, Y Combinator

    7 min read 📘 12 语块

    YC president Garry Tan demos GStack, his open-source harness that turns Claude Code into a full AI engineering team — with office hours, adversarial review, design brainstorming, and a headless browser built in.

  • Agents Need More Than a Chat — Jacob Lauritzen, CTO of Legora

    9 min read 📘 14 语块

    Why complex AI agents need high-bandwidth, persistent interfaces instead of chat boxes — trust, control, verifiability, and the future of human-agent collaboration.