Optional deterministic CI regression check for tool-using agents?

[ZackMitchell910]

Hi! I’m working on RunLedger, a small deterministic CI check for agent/tool calls. You record tool outputs once locally, then CI reuses them (no network, no secrets). It adds a tiny eval suite + workflow and is optional/removable.

If you’re open to it, what’s the best existing agent/example entrypoint to wire to for a minimal demo?

[eyurtsev]

Hi @ZackMitchell910, thanks for asking, we don't want to add this to our CI