L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Productivité · SKILL.md

matrices

matrices

Problem-solving strategies for matrices in linear algebra

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: matrices description: "Problem-solving strategies for matrices in linear algebra" allowed-tools: [Bash, Read]

Matrices

When to Use

Use this skill when working on matrices problems in linear algebra.

Decision Tree

  1. Identify Matrix Type

    • Square, symmetric, orthogonal, diagonal?
    • Check properties with sympy_compute.py matrix_type
  2. Basic Operations

    • Multiplication: sympy_compute.py matmul "A" "B"
    • Inverse: sympy_compute.py inverse "A"
    • Transpose: sympy_compute.py transpose "A"
  3. Solve Linear Systems

    • Ax = b: sympy_compute.py linsolve "A" "b"
    • Check consistency with z3_solve.py sat
  4. Decompositions

    • LU: sympy_compute.py lu "A"
    • QR: sympy_compute.py qr "A"
    • SVD: sympy_compute.py svd "A"

Tool Commands

Sympy_Inverse

uv run python -m runtime.harness scripts/sympy_compute.py inverse "[[1,2],[3,4]]"

Sympy_Det

uv run python -m runtime.harness scripts/sympy_compute.py det "[[a,b],[c,d]]"

Sympy_Linsolve

uv run python -m runtime.harness scripts/sympy_compute.py linsolve "[[1,2],[3,4]]" "[5,6]"

Cognitive Tools Reference

See .claude/skills/math-mode/SKILL.md for full tool documentation.

Skills enregistrées