leo-laptop/calculate
Use when the user asks you to calculate, compute, evaluate, or solve a math expression or equation. Triggers on arithmetic, order of operations (PEMDAS), fractions, percentages, exponents, and multi-step math problems.
84
78%
Does it follow best practices?
Impact
94%
1.00xAverage score across 5 eval scenarios
Passed
No known issues
Evaluation results
96%
4%Math Tutoring Session Transcript
Step-by-step PEMDAS evaluation format
Expression restated
100%
100%
Parentheses step labeled
100%
100%
Exponents step labeled
100%
100%
Mult/Div step labeled
100%
100%
Add/Sub step labeled
100%
100%
Intermediate results shown
100%
100%
None noted for absent steps
0%
50%
Correct answer expr 1
100%
100%
Correct answer expr 2
100%
100%
Correct answer expr 3
100%
100%
Final answer clearly stated
100%
100%
92%
-8%Retail Discount and Tax Calculator
Percentage calculations
Correct result item 1
100%
100%
Correct discount item 2
100%
100%
Correct tax item 2
100%
100%
Correct result item 3
100%
100%
Percentage step shown
100%
50%
Step-by-step work shown
100%
100%
Expression restated
100%
75%
Final answer clearly stated
100%
100%
No left-to-right error
100%
100%
PEMDAS steps labeled
100%
100%
92%
Recipe Scaling for a Bakery
Fraction arithmetic
Correct result calc 1
100%
100%
Correct result calc 2
100%
100%
Correct result calc 3
100%
100%
Common denominator shown
100%
100%
Fraction form provided
100%
100%
Decimal form provided
100%
100%
Step-by-step work shown
100%
100%
Expression restated
100%
100%
PEMDAS steps labeled
0%
0%
Final answer clearly stated
100%
100%
90%
Code Review: Math Expression Evaluator
Left-to-right equal-priority evaluation
Correct result expr 1
100%
100%
Correct result expr 2
100%
100%
Correct result expr 3
100%
100%
Correct result expr 4
100%
100%
Left-to-right rule stated
100%
100%
Step-by-step work shown
100%
100%
Expression restated
100%
100%
PEMDAS steps labeled
0%
0%
Final answer clearly stated
100%
100%
No right-to-left error
100%
100%
100%
Physics Score Adjustments
Negative numbers and negated parentheses
Correct result formula 1
100%
100%
Correct result formula 2
100%
100%
Correct result formula 3
100%
100%
Correct result formula 4
100%
100%
Negative times negative shown
100%
100%
Negated group handled correctly
100%
100%
Parentheses resolved first
100%
100%
Step-by-step work shown
100%
100%
PEMDAS steps labeled
100%
100%
Final answer clearly stated
100%
100%
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6