Agent harness
The software around a model that turns it into an agent: tool connections, memory, permissions, retries and the loop that decides what to do next.
The model supplies judgement; the harness supplies everything else. Open-source harnesses such as DeepSeek Harness, and vendor ones such as Claude Code or Codex, handle tool calling, context management, permission prompts and plugins. Most agent security properties live in the harness rather than the model, which is why a harness default such as saving approved commands to a file can leak secrets regardless of which model is running.
Related terms
AgentNon-human identityTool use (function calling)
Get the brief in your inbox
One email every morning: the stories that mattered in AI, each with a one-line "why it matters" for your role. Free, no ads, unsubscribe any time.
Published daily by AI Briefing. We never share your address.