About this session

Most of the Harness discussion is just a reprise of Context Engineering from last summer. But it's not 2025 anymore. We live in a Claude Code world, and the best way to think about a harness is Context engineering + Coding Agents = Harness. Harnesses are a magical DX because of specific features like planning mode, parallel subagents, skills, background tasks etc. But it doesn't stop there. People are shoving their harnesses in a box, making them listen to external events, giving them channels (the ability to ping its users), and a heartbeat. They are making them into Claws. And actually, harnesses _want_ to become claws, so they can take up more share of mind, suit collaboration workflows, and be available afk. I propose "Steinberger's law", a spinoff of Zawinski's law: every harness will expand until it becomes a Claw

Topics

Speaker

Sam Bhagwat

Mastra