uv-package-manager

name	uv-package-manager
description	Master the uv package manager. Triggers on: pyproject.toml, uv.lock, dependency management, pip alternatives, Dockerfile with Python, CI/CD Python setup, venv, requirements.txt migration.

UV Package Manager

Core Directives

• Always use uv instead of pip, poetry, or pip-tools for all Python dependency management.
• Never activate venvs manually (source .venv/bin/activate); use uv run <command>.
• Never generate requirements.txt unless specifically requested for a legacy system. Rely on pyproject.toml and uv.lock.

Daily Developer Workflows

Project Initialization & Python Version

uv init .              # Initialize in current directory
uv python pin 3.12     # Pin Python version (.python-version)

Adding/Removing Dependencies

uv add fastapi pydantic         # Add to dependencies
uv add --dev pytest ruff        # Add to dev-dependencies
uv remove pandas                # Remove dependency

Execution (Zero-friction venv)

uv run python src/main.py
uv run pytest tests/
uv run ruff check src/

Production & CI/CD Patterns

Lockfile Management

uv lock                         # Generate/update uv.lock
uv lock --upgrade-package httpx # Upgrade specific package

CI/CD Setup (GitHub Actions)

steps:
  - uses: actions/checkout@v4
  - name: Install uv
    uses: astral-sh/setup-uv@v2
    with:
      enable-cache: true
      cache-dependency-glob: "uv.lock"
  - name: Install dependencies
    run: uv sync --all-extras --dev
  - name: Run tests
    run: uv run pytest

Docker Best Practices (Multi-Stage)

MANDATORY: Always use cache mounts and isolate dependency installation from code copying to maximize Docker layer caching.

# BUILDER STAGE
FROM python:3.12-slim AS builder
COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/

# Optimizations for Docker mounts
ENV UV_COMPILE_BYTECODE=1 \
    UV_LINK_MODE=copy

WORKDIR /app

# 1. Install dependencies FIRST (layer caching)
RUN --mount=type=cache,target=/root/.cache/uv \
    --mount=type=bind,source=uv.lock,target=uv.lock \
    --mount=type=bind,source=pyproject.toml,target=pyproject.toml \
    uv sync --frozen --no-install-project --no-dev --no-editable

# 2. Copy code and install project
COPY src/ ./src/
RUN --mount=type=cache,target=/root/.cache/uv \
    uv sync --frozen --no-dev --no-editable

# RUNTIME STAGE
FROM python:3.12-slim AS runtime
WORKDIR /app

# Copy the isolated venv
COPY --from=builder /app/.venv /app/.venv
COPY src/ ./src/

ENV PATH="/app/.venv/bin:$PATH"
CMD ["python", "-m", "src.main"]

Anti-Patterns to Correct

❌ Anti-pattern	✅ Correct
pip install -r requirements.txt	uv sync
python -m venv .venv && source .venv/bin/activate	uv sync (creates venv automatically)
uv pip install ... inside a Dockerfile	uv sync --frozen
COPY . . before uv sync in Docker	Sync lockfiles first, copy code later