3 docs tagged with "agent-harness"

Two lessons from Codex that carry over to other tools: choose an integration depth by how much of the lifecycle you want to own, and test context compaction as a state transition.

How execution control, tools, state and verification turn model proposals into bounded work.

Using Pi 1.4.0, how a minimal harness layers its customizations, how it runs the agent loop, and why an extension is not a security boundary.