We’ve trained models on JSON schemas for “tool calls”, and then built software t... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

girvo 33 days ago | parent | context | favorite | on: The agent harness belongs outside the sandbox

We’ve trained models on JSON schemas for “tool calls”, and then built software to interpret and run those calls for the LLMs

tantalor 33 days ago [–]

> software to interpret and run those calls

Yes... That's the harness!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact