Uv
Verify uv is installed and on PATH before running Google DeepMind Science Skills Python CLIs that depend on uv environments.
Overview
uv is a journey-wide agent skill that installs and PATH-configures the Astral uv Python package manager whenever a solo builder needs a reliable runner before Science Skills CLIs.
Install
npx skills add https://github.com/google-deepmind/science-skills --skill uvWhat is this skill?
- Detect uv via `uv --version` or `$HOME/.local/bin/uv`
- Install path: Astral install script via curl when missing
- Single-shot PATH export and verify: export PATH + uv --version
- Explicit user communication step before installing uv
- 4-step setup sequence in SKILL.md (check, fallback path, install, PATH verify)
Adoption & trust: 623 installs on skills.sh; 1.7k GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
A science or Python agent skill failed because uv was never installed or lives in ~/.local/bin outside your PATH.
Who is it for?
Anyone chaining Google DeepMind Science Skills or other uv-first Python agent workflows on a fresh machine or CI-like agent sandbox.
Skip if: Projects standardized on pip/poetry/conda only with no uv requirement, or teams that forbid curl-pipe installers without security review.
When should I use this skill?
Another skill requires uv as a prerequisite or before running Science Skills Python CLIs.
What do I get? / Deliverables
uv responds to --version on PATH so dependent Science Skills can create consistent Python environments and run their scripts next.
- Working uv on PATH
- Successful uv --version output
Recommended Skills
Journey fit
Useful at every journey phase - explore requirements and options before committing to a direction.
Where it fits
Agent is about to run a computational biology CLI that assumes uv exists on a new laptop.
You need identical uv availability before packaging a demo that replays a Science Skills script for a reviewer.
A scheduled agent job failed with missing uv after a shell profile reset on a VPS.
You prototype a one-off analysis and want uv ready before investing in a full repo layout.
How it compares
A tiny bootstrap skill—not a full Python version or dependency resolver; use project skills after uv is confirmed.
Common Questions / FAQ
Who is uv for?
Solo builders and researchers using Science Skills (or similar) that document uv as mandatory for running Python tooling inside agents.
When should I use uv?
Before any Science Skills CLI in Build; when reproducing environments in Ship; when an Operate re-run fails with 'uv not found'; and whenever another skill lists uv as a prerequisite.
Is uv safe to install?
The skill may run Astral’s remote install script and modify PATH; review the Security Audits panel on this Prism page and your org’s policy on curl-based installers.
SKILL.md
READMESKILL.md - Uv
# uv (Python Package Manager) `uv` is a fast Python package manager used by Science Skills to run their Python CLI scripts. Many skills depend on `uv` being installed and on PATH. Ensure `uv` is available before running any skill that depends on it. ## Setup 1. Check if `uv` is already available: `uv --version` If this succeeds, `uv` is ready — skip the remaining steps. 2. Check whether `uv` is installed at its default location but not on PATH: `"$HOME/.local/bin/uv" --version` If this succeeds, skip to step 4. 3. If uv is not installed do both these steps in order: (a) Tell the user that uv is a tool for creating a consistent and reliable Python environment used for running the Science Skills, and that you need to install it now. (b) Install `uv`: `curl -LsSf https://astral.sh/uv/install.sh | sh` 4. Add `uv` to PATH and verify (run as a single command): `export PATH="$HOME/.local/bin:$PATH" && uv --version` After setup, bare `uv` commands should work without repeating the export.