python-with-uv

Python Development with UV

Overview

All Python work on this system uses the uv package manager exclusively. UV handles virtual environments automatically - no manual activation, no pip, no confusion.

Core principle: UV replaces both pip and manual venv management. One tool, zero friction.

When to Use

Use this workflow when:

• Setting up a new Python project
• Installing Python packages
• Running Python scripts or commands
• Managing project dependencies
• Working with virtual environments

Do NOT use for:

• Non-Python projects
• System Python packages (use system package manager)

Quick Reference

Task	Command	Notes
Initialize project	uv init --bare && uv venv --seed --prompt " 🌞 "	Sets up bare project with seeded venv
Add dependency	uv add <package> && uv sync	Always sync after adding
Remove dependency	uv remove <package> && uv sync	Always sync after removing
Run Python script	uv run python script.py	No venv activation needed
Run any command	uv run <command>	Runs in project's venv context
Install all deps	uv sync	Syncs pyproject.toml to venv
Update dependencies	uv lock --upgrade	Updates lock file

Core Rules

✅ ALWAYS

• Use uv for ALL Python operations
• Run commands with uv run prefix
• Sync after adding/removing packages: uv add pkg && uv sync
• Work from correct project directory (UV is directory-aware)

❌ NEVER

• Use pip or pip3 (blocked in permissions)
• Run python or python3 directly (blocked in permissions)
• Manually activate virtual environments (source venv/bin/activate)
• Use virtualenv or venv commands

Common Mistakes

Mistake	Why It's Wrong	Correct Approach
pip install requests	Uses pip instead of uv	uv add requests && uv sync
python script.py	Bypasses UV's venv handling	uv run python script.py
uv add pandas (without sync)	Dependencies not installed	uv add pandas && uv sync
source .venv/bin/activate	Manual venv management unnecessary	Just use uv run
Running from wrong directory	UV can't find pyproject.toml	cd to project root first

Working Directory Requirement

UV requires you to be in the correct project directory because it looks for pyproject.toml:

# ❌ Wrong - not in project directory
cd ~
uv run python my_project/script.py  # Won't find pyproject.toml

# ✅ Correct - in project directory
cd ~/my_project
uv run python script.py  # Finds pyproject.toml

Workflow Examples

Starting a New Project

# Navigate to projects directory
cd ~/projects/my-new-project

# Initialize with bare template and seeded venv
uv init --bare && uv venv --seed --prompt " 🌞 "

# Add dependencies
uv add pandas numpy && uv sync

# Create and run script
uv run python analyze.py

Adding Packages to Existing Project

# Navigate to project
cd ~/projects/my-project

# Add package and sync in one command
uv add requests && uv sync

# Verify it works
uv run python -c "import requests; print(requests.__version__)"

Running Scripts

# Simple script execution
uv run python main.py

# Script with arguments
uv run python process.py --input data.csv --output results.json

# Interactive Python shell with project dependencies
uv run python

Red Flags - STOP and Use UV Instead

If you catch yourself about to:

• Type pip install
• Run python directly
• Create/activate a virtualenv manually
• Wonder "which Python am I using?"

STOP. Use UV commands instead.

Benefits Over pip/venv

• No activation needed: uv run handles venv automatically
• Faster installs: UV is written in Rust, significantly faster than pip
• Simpler workflow: One tool instead of pip + venv + virtualenv
• Better dependency resolution: UV has superior conflict resolution
• Lockfile support: Reproducible environments via uv.lock

Troubleshooting

"Command not found: uv"

• UV not installed. Install with: brew install uv

"No pyproject.toml found"

• Not in project directory. Run cd to project root or initialize with uv init --bare.

"Package not found after uv add"

• Forgot to sync. Always use: uv add package && uv sync

"Wrong Python version"

• UV uses Python from pyproject.toml. Specify version: uv python pin 3.11