Agent Harness

Agent Harness is an execution layer for AI coding workflows.

It is built around one simple operating model:

1. use a strong supervisor model for planning and review
2. delegate bounded tasks to cheaper or local workers
3. verify results deterministically
4. keep the operator in control when approval, failure, or recovery matters

What it is not

It is not trying to be a general chat UI, a hosted control plane, or a vague “agent platform.”

The current product path is narrower and more useful:

• standalone OSS first
• reliable local and mixed-model execution
• evidence-rich artifacts
• benchmarkable routing and retrieval improvements

Product priorities

The current roadmap focuses on four phases:

• Phase 12: standalone adoption and release hardening
• Phase 13: reliability and evidence-first execution quality
• Phase 14: retrieval, routing, and benchmark quality
• Phase 15: OSS-first expansion toward team-ready operation