Glossary · Phase 10 · Agent harnesses
Test harness
Code that runs other code under controlled conditions and checks the results.
Try it now
Each reply is one model call. Two ask for tools, so the loop runs them and goes round again; the third ends the turn.
Learn it properly
What an agent harness is · The software around a model that turns it into an agent. About 30 minutes.
Other terms from the same lesson
- agent harness: The software around a model that turns it into an agent: the loop, tools, permissions, context management, memory and logging.
- Claude Agent SDK: A library that packages Claude Code's harness: built-in file, shell and web tools, permissions, hooks, subagents and sessions, run on your own machines.
- managed agents: A hosted option where Anthropic runs the agent loop and a sandbox container for each session.
- scripted model: A stand-in for a model that returns prepared replies in order, so harness code can be tested quickly and repeatably.
- tool runner: A helper in the Anthropic SDKs that runs the agent loop over tools you define.
Never trained an AI?
Teach a small one to write, right in your browser. It takes about 20 seconds and needs no account.