OrkasOrkas
DownloadGitHub

calculator

Run restricted local Python calculations for expressions, arbitrary precision, matrices, statistics, optimization, integration, and signal transforms; For: "calculate this expression", "verify this math result with calculator", "batch-calculate these formulas"; Triggers: calculator, math calculation, expression evaluation, formula verification, numerical computation, matrix calculation, statistics, calc

Category: General · Version 1.0.0

Get this in the Orkas app

What this skill does

Calculator

When To Use

  • The user wants a deterministic numeric calculation rather than a conceptual math explanation.
  • The task involves expression evaluation, high-precision constants or large numbers, matrices, descriptive statistics, correlations, regressions, distribution pdf/cdf values, scalar optimization, roots, definite integrals, simple ODE samples, FFT/IFFT, or convolution.
  • The user asks to verify arithmetic, formulas, generated answers, or intermediate numeric results.

Do not use for K12 tutoring, grading, curriculum-aligned explanations, or error diagnosis; those require a dedicated learning-support workflow. Do not use for open-ended mathematical modeling; keep modeling assumptions and model selection in a dedicated modeling workflow when available.

How To Call

Use the bundled script through Orkas:

$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- eval "2**10 + sin(pi/4)"

Choose the subcommand by computation type:

| Need | Subcommand | Example | | --- | --- | --- | | Daily arithmetic, trig, logs, powers | eval | eval "sin(pi/4) + sqrt(2)" | | Arbitrary precision, factorial, gamma, zeta | precision | precision "factorial(100)" --precision 50 | | Matrix determinant, inverse, solve, eigenvalues | matrix | matrix det --matrix "[[1,2],[3,4]]" | | Descriptive stats, correlation, regression, pdf/cdf | stats | stats describe --data "[1,2,3,4,5]" | | Scalar minimization, maximization, roots | optimize | optimize root --expr "x**3 - 2*x - 5" --bounds "[1,3]" | | Definite integrals or simple ODE samples | integrate | integrate definite --expr "sin(x)" --bounds "[0,pi]" | | FFT, IFFT, convolution | transform | transform fft --data "[1,0,1,0]" |

Core examples:

$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- eval "2**10 + sin(pi/4)"
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- eval "pi" --precision 50
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- precision "factorial(1000)" --precision 50
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- matrix inverse --matrix "[[1,2],[3,4]]"
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- matrix solve --matrix "[[2,1],[1,-1]]" --matrix2 "[5,1]"
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- stats regression --data "[1,2,3]" --data2 "[2,4,6]"
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- optimize minimize --expr "x**2 + 2*x + 1" --bounds "[-10,10]"
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- integrate definite --expr "sin(x)" --bounds "[0,pi]"
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- transform convolve --data "[1,2,3]" --data2 "[4,5]"

Workflow

  1. Translate the user's calculation request into the smallest matching subcommand.
  2. Run the script before reporting numeric answers when exact numeric verification matters.
  3. Include the expression, subcommand, and result in the response when the user may need to audit the calculation.
  4. For high precision, large factorials, special functions, or results sensitive to floating-point rounding, use precision instead of eval.
  5. For expressions starting with -, pass the value with equals syntax, such as --expr="-x**2+4", so argument parsing treats it as a value.
  6. If the script errors because a dependency is missing, report the missing package and either compute by another reliable available method or ask before installing anything.

Output Shape

The script prints:

=== <subcommand> ===
<result>

When responding to the user, keep the answer compact:

Result: ...
Check: ...

For multi-step verification:

**Calculation**
- Command:
- Result:
- Interpretation:

Dependencies

  • Python 3.
  • Basic scalar eval can run with Python stdlib math functions.
  • numpy is required for matrix, statistics, transforms, and array-oriented expression behavior.
  • mpmath is required for arbitrary precision.
  • scipy is required for distribution functions, optimization, and integration.

Limits And Safety

  • Expressions are evaluated with restricted namespaces for math functions and constants, but still treat user expressions as calculation input only.
  • eval uses float64-style numeric behavior unless --precision is supplied.
  • Matrix and data inputs use Python literal syntax such as [[1,2],[3,4]] or [1,2,3].
  • This is not a symbolic algebra system. For proofs, derivations, or curriculum-aligned explanation, use reasoning or a dedicated learning-support workflow alongside calculator verification.

Files in this skill

  • _meta.json
  • scripts/calc.py
  • SKILL.md

Related Skills

How to use in Orkas

Open the Orkas desktop app, go to the marketplace, and install this item with one click. Don't have Orkas yet? Download Orkas.

用本地 Python 脚本进行受限数学表达式、任意精度、矩阵、统计、优化、积分和信号变换计算；适合"帮我算这个表达式""用 calculator 验证数学结果""批量计算这些公式"；触发词：计算器、数学计算、表达式求值、公式验证、数值计算、矩阵计算、统计计算、calculator、calc

分类：通用 · 版本 1.0.0

在 Orkas 应用中获取

该技能的功能

Calculator

When To Use

  • The user wants a deterministic numeric calculation rather than a conceptual math explanation.
  • The task involves expression evaluation, high-precision constants or large numbers, matrices, descriptive statistics, correlations, regressions, distribution pdf/cdf values, scalar optimization, roots, definite integrals, simple ODE samples, FFT/IFFT, or convolution.
  • The user asks to verify arithmetic, formulas, generated answers, or intermediate numeric results.

Do not use for K12 tutoring, grading, curriculum-aligned explanations, or error diagnosis; those require a dedicated learning-support workflow. Do not use for open-ended mathematical modeling; keep modeling assumptions and model selection in a dedicated modeling workflow when available.

How To Call

Use the bundled script through Orkas:

$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- eval "2**10 + sin(pi/4)"

Choose the subcommand by computation type:

| Need | Subcommand | Example | | --- | --- | --- | | Daily arithmetic, trig, logs, powers | eval | eval "sin(pi/4) + sqrt(2)" | | Arbitrary precision, factorial, gamma, zeta | precision | precision "factorial(100)" --precision 50 | | Matrix determinant, inverse, solve, eigenvalues | matrix | matrix det --matrix "[[1,2],[3,4]]" | | Descriptive stats, correlation, regression, pdf/cdf | stats | stats describe --data "[1,2,3,4,5]" | | Scalar minimization, maximization, roots | optimize | optimize root --expr "x**3 - 2*x - 5" --bounds "[1,3]" | | Definite integrals or simple ODE samples | integrate | integrate definite --expr "sin(x)" --bounds "[0,pi]" | | FFT, IFFT, convolution | transform | transform fft --data "[1,0,1,0]" |

Core examples:

$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- eval "2**10 + sin(pi/4)"
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- eval "pi" --precision 50
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- precision "factorial(1000)" --precision 50
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- matrix inverse --matrix "[[1,2],[3,4]]"
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- matrix solve --matrix "[[2,1],[1,-1]]" --matrix2 "[5,1]"
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- stats regression --data "[1,2,3]" --data2 "[2,4,6]"
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- optimize minimize --expr "x**2 + 2*x + 1" --bounds "[-10,10]"
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- integrate definite --expr "sin(x)" --bounds "[0,pi]"
$ORKAS_NODE $ORKAS_PC_DIR/bin/run-skill.cjs calculator calc -- transform convolve --data "[1,2,3]" --data2 "[4,5]"

Workflow

  1. Translate the user's calculation request into the smallest matching subcommand.
  2. Run the script before reporting numeric answers when exact numeric verification matters.
  3. Include the expression, subcommand, and result in the response when the user may need to audit the calculation.
  4. For high precision, large factorials, special functions, or results sensitive to floating-point rounding, use precision instead of eval.
  5. For expressions starting with -, pass the value with equals syntax, such as --expr="-x**2+4", so argument parsing treats it as a value.
  6. If the script errors because a dependency is missing, report the missing package and either compute by another reliable available method or ask before installing anything.

Output Shape

The script prints:

=== <subcommand> ===
<result>

When responding to the user, keep the answer compact:

Result: ...
Check: ...

For multi-step verification:

**Calculation**
- Command:
- Result:
- Interpretation:

Dependencies

  • Python 3.
  • Basic scalar eval can run with Python stdlib math functions.
  • numpy is required for matrix, statistics, transforms, and array-oriented expression behavior.
  • mpmath is required for arbitrary precision.
  • scipy is required for distribution functions, optimization, and integration.

Limits And Safety

  • Expressions are evaluated with restricted namespaces for math functions and constants, but still treat user expressions as calculation input only.
  • eval uses float64-style numeric behavior unless --precision is supplied.
  • Matrix and data inputs use Python literal syntax such as [[1,2],[3,4]] or [1,2,3].
  • This is not a symbolic algebra system. For proofs, derivations, or curriculum-aligned explanation, use reasoning or a dedicated learning-support workflow alongside calculator verification.

技能包含的文件

  • _meta.json
  • scripts/calc.py
  • SKILL.md

相关技能

如何在 Orkas 中使用

打开 Orkas 桌面应用，进入市场，一键安装此项。还没有 Orkas？ 下载 Orkas.