uv
how to run python files and mangage dependencies using UV package manager (pip replacement)
When & Why to Use This Skill
This Claude skill optimizes Python development workflows by integrating the UV package manager, a high-performance alternative to pip, pip-tools, and virtualenv. It enables faster dependency resolution, automated environment synchronization, and standardized script execution to ensure project reproducibility and eliminate common environment-related errors.
Use Cases
- Rapid Project Initialization: Quickly set up a consistent Python development environment by synchronizing all dependencies from a pyproject.toml file using 'uv sync'.
- Efficient Dependency Management: Seamlessly add, update, or remove production and development packages while maintaining a clean and locked dependency graph.
- Isolated Script Execution: Run Python scripts and tools within a managed environment using 'uv run' to prevent version conflicts and ensure the correct virtual environment is always active.
- Modernizing Python Workflows: Transitioning legacy projects from pip/virtualenv to UV to significantly reduce build times and simplify CI/CD pipeline configurations.
| name | "uv" |
|---|---|
| description | "how to run python files and mangage dependencies using UV package manager (pip replacement)" |
You are working with a Python project that uses UV package manager. UV is a fast package manager serving as a drop-in replacement for pip, pip-tools, and virtualenv.
Core Commands
Project Setup
uv sync # Install all default AND dev dependencies from pyproject.toml
Running Commands
uv run python script.py # Run Python with project dependencies
Adding Dependencies
uv add package-name # Add to [project.dependencies]
uv add --dev package-name # Add to [dependency-groups.dev]
CRITICAL: Anti-Patterns
NEVER use bare python
# FORBIDDEN:
python ...
python3 ...
INSTEAD: Always use uv run python ...
NEVER use pip
# FORBIDDEN:
uv pip ...
pip ...
python -m pip ...
uv run python -m pip ...
INSTEAD: Use uv add and uv sync
Instructions
When working with Python files in this project:
- Always use
uv run pythoninstead of barepythoncommands - Install dependencies with
uv add, never withpip - Sync dependencies with
uv syncafter modifying pyproject.toml - Run scripts and tools through
uv runto ensure correct environment