What evals decide
Evals grade every agent's work and gate every new version, and never change what waits for the founder.
Evals grade each Agent's work and decide which version of it ships. What waits for the founder is set by approvals, by action class and amount, and no eval score changes it.
The record
Grades come from three sources: the Role's offline eval cases, Evaluator verdicts on live Actions, and the founder's corrections. Every edit, refusal or redirect of an Action becomes a new eval case. The Evals record keeps every grade.
A falling pass rate opens a Task for the seat's Role Review. The review ends in a Decision: change the Role contract, tighten an approval, or seat another Worker.
New versions earn their way in
Skills, prompts, Role policies and Mandates are versioned definitions with their own eval suites. A new version must score no worse than the current one offline before it touches a customer. It then runs as an Experiment against the Startup's one Metric, with guardrails.
A version that fails never ships. No Agent edits the Metrics, Evaluators or guardrails that grade it.
One model for every Startup
Every Startup runs the same way: agents act, one-way decisions wait their delay windows, and watchers can stop them. A founder adds approvals by class and amount. Every Startup shares the same Limits, spend states and legal floor.
In practice
A Services-as-Software bookkeeping firm for dental practices approves spend above $100, so refunds under $100 run without the founder. In month three, one refund goes to a closed bank account and an Evaluator fails it. The case joins the suite, and the next Skill version must pass it before it ships.