uv-python-workflow

Original：🇺🇸 English

Translated

Use uv as the exclusive Python package and environment manager. Use this skill when installing packages, syncing environments, running scripts, writing Dockerfiles, or doing any Python dependency management — never pip, pip-compile, or virtualenv directly.

10installs

Sourcemitodl/agent-kit

Added on2026-05-23

NPX Install

npx skill4agent add mitodl/agent-kit uv-python-workflow

SKILL.md Content

View Translation Comparison →

uv Python Workflow

All Python dependency and environment management in this codebase uses uv
exclusively.

Package management

Add dependencies:
```
uv add <package>
```
Remove dependencies:
```
uv remove <package>
```
Install/sync the current project:
```
uv sync
```
Run a command inside the project environment:
```
uv run <command>
```
Never use
```
pip install
```
,
```
pip-compile
```
, or bare
```
python -m venv
```
.

Workspace layout

Projects may be organized as a

uv

workspace. Code locations managed by

dg

do not use

uv

workspaces — they rely on the

dg

workspace instead.

Docker / container images

When building a Python app inside a Docker image, create a relocatable virtual environment so it works after the image layers are assembled:

dockerfile

RUN uv venv --relocatable /app/.venv
RUN uv sync --frozen --no-dev

Do not mount

ol-orchestrate-lib

(or any shared library) as a volume at runtime. Shared libraries are installed as build-time dependencies only.

Running pre-commit and type checks

bash

uv run pre-commit run --all-files
uv run mypy <package>