The Evolution of the Agent Harness
Latent Space essay argues late-2025 agent gains came from models and harnesses maturing together, with harness logic absorbed into model weights.
The piece defines the agent harness as everything beyond model weights—tools, context, memory, guardrails—and charts its evolution from ReAct prompting (October 2022) through AutoGPT's premature autonomy, Cursor/Copilot's human-in-the-loop retreat, and Devin's roughly 15% success rate, to o1's capability overhang and Claude Code's February 2025 terminal agent with permission rules. It argues the Christmas 2025 jump cited by Transformer co-inventor Lukasz Kaiser reflected model and harness curves crossing, and that remaining harnesses will serve human attention rather than the model.
- Agent harness = environment, tools, memory, and guardrails around weights
- AutoGPT-style full autonomy amplified errors below a reliability threshold
- Claude Code abandoned the IDE, giving the model bash and file access
- Models keep absorbing harness functions into weights as capability grows
- Organizations
- Latent Space
From the original article
9 min read at latent.space
Models keep absorbing the harness into their weights — soon, it will be a harness for human attention rather than for the model.
By Dan McAteer · latent.space
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.