Your agent picks one of two options. Can you test that choice?

Automated excerpt

You put both in the prompt, ask for a pick, and get back something confident. One tool, narrow contract: confidence: "83. 3%" // how far apart the two were judged It's a separate call, not an aside. High confidence means the two options were judged clearly apart.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.