Skip to content

devskills-cli

An interactive CLI to scaffold AI/ML production projects using the uv package manager. No flags — just prompts.

Why devskills-cli?

Starting an AI/ML API project involves the same repetitive setup every time — folder structure, .gitignore, Docker config, Claude Code integration. devskills-cli does all of that in under 60 seconds.

pip install devskills-cli
dev start

That's it. Answer four questions, get a production-ready project.

What it does

Scaffolds a complete project structure interactively:

  • Pick a project name (or scaffold in the current directory)
  • Choose a structure: AI/ML, API, Minimal, or None
  • Add optional features: Docker, Claude Code
  • Create a .venv immediately

Manages Claude Code skills from skills.sh:

  • dev skills find <keyword> — search and install
  • dev skills install <owner/repo> — install directly
  • dev skills list — browse top skills

Quick example

dev start
# > Project name: my-ai-api
# > Structure: AI / ML
# > Features: [x] Claude  [x] Docker
# > Create .venv? Yes

cd my-ai-api
uv sync

Your project is ready. Claude Code knows the layout. Docker is configured. Start building.

Requirements

Tool Version Purpose
Python 3.10+ Runtime
uv latest Project init and virtualenv
Node.js / npx any dev skills commands only