When two other AIs check the answer.
The explanation says what happens to their findings: disagreements are marked in the answer, additions are appended and attributed. A crosscheck whose output is silently merged is a crosscheck nobody can weigh.
always states the real cost - about twenty seconds and a few cents per answer. A safeguard whose price is hidden gets switched off the first time somebody notices the bill, which is the worst moment to learn about it.
import { SecondOpinionSetting } from '@redrob-labs/ui';
| Prop | Type | Notes |
|---|---|---|
options |
OpinionMode[] |
|
value |
string |
|
defaultValue |
string |
|
title |
React.ReactNode |
|
lede |
React.ReactNode |
|
label |
string |
|
foot |
React.ReactNode |
|
className |
string |
|
onChange |
(value: string, mode: OpinionMode) => void |