Foreman: a supervisor for coding agents

Foreman watches coding agents and uses Jev to continuously assess progress, completeness, tests, drift and verification, intervening when needed.

Notes

Python asyncio runtime: a Codex worker runs in one loop while Jev scores nine dimensions (completeness, test sufficiency, requirements, stuck, drift and more) in parallel and a deterministic policy continues, retries, verifies or escalates; default ready threshold 0.85; MIT.

source-backed — Public repository, docs or live artifact. About this label

Destinations

Try a walkthrough

• Route the next step of an agent →

  Let Jev choose a proposed next step, then apply an explicit allow, review or fallback policy before any tool runs.

Verification levels

Report a problem →

Filed under

• #coding
• #agents
• #supervision

In collections

• Agent harnesses →
• Coding workflows →