claude-awesome-stack

Installable stack packs for Claude Code -- production-ready configurations of skills, hooks, and project instructions for domain-specific development.

Unlike link-collection "awesome lists", this repo gives you working setups you copy into your project for an immediate, opinionated Claude Code experience.

How It Works

graph LR
    A[claude-awesome-stack] -->|copy| B[Your Project]
    subgraph "Stack Pack"
        C[CLAUDE.md]
        D[skills/]
        E[hooks/]
        F[settings.json]
    end
    subgraph "Your Project"
        G[CLAUDE.md]
        H[.claude/skills/]
        I[.claude/hooks/]
        J[.claude/settings.json]
    end
    C --> G
    D --> H
    E --> I
    F --> J

Loading

Quick Start

# 1. Clone
git clone https://github.com/giacomogaglione/claude-awesome-stack.git

# 2. Copy a stack into your project
cp -r claude-awesome-stack/stacks/python-ml/CLAUDE.md ./CLAUDE.md
mkdir -p .claude/skills .claude/hooks
cp -r claude-awesome-stack/stacks/python-ml/skills/* .claude/skills/
cp -r claude-awesome-stack/stacks/python-ml/hooks/* .claude/hooks/
cp claude-awesome-stack/stacks/python-ml/settings.json .claude/settings.json

# 3. Start Claude Code
claude

Available Stacks

Stack	Status	Description
Python/ML	Ready	ML conventions, debugging skills, auto-formatting hooks
Frontend	Coming soon	React/Next.js development

Python/ML Stack

Component	What It Does
CLAUDE.md	ML conventions: tensor shapes, dtype checks, anti-patterns, testing patterns
ml-debug skill	Systematic debugging: shapes, NaN propagation, gradients, devices, dtypes
data-explore skill	Dataset profiling: schema, distributions, missing values, outliers, correlations
notebook-refactor skill	Extract Jupyter notebooks into tested Python modules
python-lint.sh hook	Auto-format with ruff + black on every file edit
type-check.sh hook	pyright/mypy type checking after edits
test-guard.sh hook	Run relevant pytest tests when source files change
pyproject.toml template	Dev/ML/torch dependency groups with ruff, black, pyright config
.env.example template	Common ML environment variables

Hook Execution Flow

sequenceDiagram
    participant C as Claude Code
    participant H as Hooks
    participant T as Tools

    C->>C: Edit Python file
    C->>H: PostToolUse event (JSON)
    H->>T: ruff check --fix
    H->>T: black format
    H->>T: pyright type-check
    H->>T: pytest (matching tests)
    H-->>C: Results fed back to conversation

Loading

Curated Resources

Hand-picked tools and patterns for Claude Code:

Resource	Description
MCP Servers	Categorized MCP servers for data, cloud, monitoring, and more
Skills	Community skills reviewed for quality and compatibility
Hook Recipes	Ready-to-use hook patterns for linting, testing, security, and productivity

Documentation

• Installation Guide -- step-by-step setup
• Architecture -- how stacks work and why
• Creating Stacks -- contributor guide for new stacks

Contributing

See CONTRIBUTING.md for how to submit stacks, skills, hook recipes, or curated list entries.

License

MIT