Claude Code skill for writing idiomatic Apple MLX code on Apple Silicon. Covers lazy evaluation, array indexing, neural networks, compilation, memory management, and PyTorch migration.
npx claudepluginhub joshuarweaver/cascade-code-languages-python --plugin ettrickshepherd-mlx-dev-skillA Claude Code skill for writing correct, idiomatic Apple MLX code on Apple Silicon.
In Claude Code, run:
/plugin marketplace add tkwn2080/mlx-dev-skill
The skill will be automatically available after adding the marketplace.
Clone and copy to your personal skills directory:
git clone https://github.com/tkwn2080/mlx-dev-skill.git
cp -r mlx-dev-skill/skills/mlx-dev ~/.claude/skills/
Then restart Claude Code.
When working with MLX, Claude will automatically:
mx.eval() at loop boundariesmx.array, slices create copies)__call__() not forward() for nn.Modulemx.compile()at[] syntax, gather/scattermx.compile() patterns, state capture, shapeless modevalue_and_grad, custom vjp, control flowmlx-dev-skill/
├── .claude-plugin/
│ └── plugin.json # Plugin metadata
├── skills/
│ └── mlx-dev/
│ ├── SKILL.md # Main skill entry point
│ ├── references/
│ │ ├── array-indexing.md
│ │ ├── compilation.md
│ │ ├── dtypes.md
│ │ ├── error-decoder.md
│ │ ├── gradients.md
│ │ ├── memory-management.md
│ │ ├── neural-networks.md
│ │ ├── pytorch-migration.md
│ │ └── random.md
│ └── scripts/
│ └── check_memory.py
├── README.md
└── LICENSE
The skill includes a memory debugging script:
# Show current memory stats
uv run python ~/.claude/skills/mlx-dev/scripts/check_memory.py
# Monitor continuously
uv run python ~/.claude/skills/mlx-dev/scripts/check_memory.py --watch
# Log to CSV for analysis
uv run python ~/.claude/skills/mlx-dev/scripts/check_memory.py --watch --log memory.csv
uv add mlx)MIT
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Creative skill for generating algorithmic and generative art. Produces visual designs using mathematical patterns, fractals, and procedural generation.
Frontend design skill for UI/UX implementation
Humanise text and remove AI writing patterns. Detects and fixes 24 AI tell-tales including inflated language, promotional tone, AI vocabulary, filler phrases, sycophantic tone, and formulaic structure.
Expert guidance for Next.js Cache Components and Partial Prerendering (PPR). Proactively activates in projects with cacheComponents: true, providing patterns for 'use cache' directive, cacheLife(), cacheTag(), cache invalidation, and parameter permutation rendering.