AI Dictionary › Grace
The score in Grace is the numerical rating from 0 to 100 assigned by the AI judge to each prompt. It is the weighted average of scores across the 7 dimensions of the evaluation rubric (Role Assignment, Context, Specific Objective, Constraints, Output Format, Audience, Token Efficiency); the weight of each dimension changes from scenario to scenario, following that specific case's rubric (see the "Evaluation rubric" entry).
The score directly determines the XP earned: XP = base × (score ÷ 100) × 2. A score of 50 produces 50% of the maximum XP, a score of 100 produces 100%. Active bonuses (improvement, first area, streak) are added after the multiplier.
The score is also the basis for calculating the user's percentile relative to every other attempt on the same scenario. A score in the top 5% (percentile ≥ 95) automatically triggers the credit bonus. Grace's scoring system turns the evaluation of prompt quality into a structured, repeatable rating against a public rubric: not an absolute measure, but a consistent, comparable reference across attempts.
After each scenario, your score appears immediately with the breakdown by dimension. You can see how you performed on Role Assignment, Context, Specific Objective, Constraints, Format, Audience and Token Efficiency. This granular feedback is what makes Grace an effective learning tool rather than just a scoring system.
From our network
Magellano GPS: Fleet Tracking Made Simple
Real-time GPS tracking, remote engine lock, fuel and CO₂ reporting for your fleet.
Visit magellanogps.com →From the Agora Intelligence blog
📱 Download the Android app (beta) iOS coming soon
Say what you mean. Get what you need.
Grace Certified, the AI coach that trains and certifies your prompt engineering, by Agora Intelligence.