AgentSkillsCN

uv-python-env

在工作空间中管理基于 uv 的 Python 环境。当 Python 或相关软件包缺失、uv 虚拟环境未创建，或是你需要初始化或同步依赖项、以 uv 运行 Python 时，这一功能将为你保驾护航。

SKILL.md
--- frontmatter
name: uv-python-env
description: Manage uv-based Python environments in a workspace. Use when Python or packages are missing, when a uv venv is absent, or when you need to initialize or sync dependencies with uv and run Python via uv.

UV Python Environment

Core workflow

  • •Work from the workspace root.
  • •If no pyproject.toml exists, initialize the project with uv init ..
  • •If pyproject.toml exists and dependencies are defined, create or repair the env with uv sync --no-cache --upgrade.
  • •Use uv add <pkg> for runtime dependencies and uv add --dev <pkg> for dev tools (e.g., ipython, rich, see).

Activate and run

  • •Activate the venv with source /path/to/workspace/.venv/bin/activate.
  • •Prefer uv run python <file.py> to run scripts.
  • •If a Python snippet is needed, write it to a .py file first and then run it. Avoid uv run python -c "<multiline>".
  • •When writing a helper script, avoid argparse CLI scaffolding. Prefer small, local, low-coupling functions that take arguments, and call them explicitly under if __name__ == "__main__": without a main() function when possible.

Troubleshooting

  • •If Python is not found, activate the venv or run with uv run.
  • •If a package is missing, add it with uv add (or uv add --dev for dev-only tools), then retry.