# Claude Awesome Stack

> Installable stack packs for Claude Code — production-ready skills, hooks, and project configs for domain-specific development

- **Type:** MCP server
- **Install:** `agentstack add mcp-giacomogaglione-claude-awesome-stack`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [giacomogaglione](https://agentstack.voostack.com/s/giacomogaglione)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [giacomogaglione](https://github.com/giacomogaglione)
- **Source:** https://github.com/giacomogaglione/claude-awesome-stack

## Install

```sh
agentstack add mcp-giacomogaglione-claude-awesome-stack
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# 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

```mermaid
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
```

## Quick Start

```bash
# 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](stacks/python-ml/) | Ready | ML conventions, debugging skills, auto-formatting hooks |
| [Frontend](stacks/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

```mermaid
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
```

## Curated Resources

Hand-picked tools and patterns for Claude Code:

| Resource | Description |
|----------|-------------|
| [MCP Servers](curated/mcp-servers.md) | Categorized MCP servers for data, cloud, monitoring, and more |
| [Skills](curated/skills.md) | Community skills reviewed for quality and compatibility |
| [Hook Recipes](curated/hooks.md) | Ready-to-use hook patterns for linting, testing, security, and productivity |

## Documentation

- [Installation Guide](docs/install.md) -- step-by-step setup
- [Architecture](docs/architecture.md) -- how stacks work and why
- [Creating Stacks](docs/creating-stacks.md) -- contributor guide for new stacks

## Contributing

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

## License

[MIT](LICENSE)

## Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [giacomogaglione](https://github.com/giacomogaglione)
- **Source:** [giacomogaglione/claude-awesome-stack](https://github.com/giacomogaglione/claude-awesome-stack)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** yes
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/mcp-giacomogaglione-claude-awesome-stack
- Seller: https://agentstack.voostack.com/s/giacomogaglione
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.