AWS bolts together open source agent harness, says it sips fewer tokens than rivals
Strands claims near-parity with Claude Code and Codex on benchmarks, though it only raced coding agents and marked its own homework
Amazon Web Services (AWS) has launched a new open source agent harness called Strands, which it claims delivers comparable benchmark results to rivals while consuming 28% fewer tokens. The harness, built on AWS' Strands Harness SDK, is designed to be plug-and-play, allowing users to integrate it into their preferred AI provider with just one line of Python or TypeScript code.
AWS emphasizes that Strands was built as a general-purpose agent, not specifically for coding tasks, which differentiates it from other harnesses that focus on coding agents like Claude Code, Codex, oh-my-pi, OpenCode, and DeepSeek Harness. The company attributes its performance to features like default prompt caching, context management, and intelligent handling of context windows.
Amazon plans to publish a detailed paper with benchmark results, but the Strands harness is immediately available for users to experiment with, offering a Python or TypeScript library, CLI, and pre-configured skills file for easy integration of custom tools.
Written by urgent.news from The Register's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
