skills.rest
jerrylin96
jerrylin96COMMUNITY
Explore all skills by this author →
WebsiteGitHub

managing-python-dependencies

Detects project dependency managers and installs Python packages with the correct tooling.

9|Updated Jul 8, 2026
Search Tags:#pip#dependency management#python#uv#poetry#virtual environment#venv
One-click install
npx skills add https://github.com/jerrylin96/dotgemini --skill managing-python-dependencies-jerrylin96
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: managing-python-dependencies
Source: https://github.com/jerrylin96/dotgemini/tree/main/skills/managing-python-dependencies
Command: npx skills add https://github.com/jerrylin96/dotgemini --skill managing-python-dependencies-jerrylin96

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? It prevents accidental global pip install calls and conflicts caused by overriding a project's established dependency manager, ensuring Python packages are installed through the tooling the project already uses. ## Core Features & Use Cases - Dependency Manager Detection: Scans the workspace for signal files (uv.lock, pyproject.toml, Pipfile, environment.yml, requirements.txt) in priority order to identify the correct tool (uv, Poetry, Pipenv, Conda, or venv + pip). - Safe Default Workflow: Falls back to an isolated .venv with explicit .venv/bin/pip paths and requirements.txt freezing when no manager is detected. - macOS CPU-Only Review Environments: Optionally provisions hashed, per-workspace virtual environments that filter out CUDA/NVIDIA dependencies for local testing and linting on macOS. - Use Case: When asked to add httpx to a repository containing a uv.lock file, run uv add httpx instead of a global pip install httpx. ## Quick Start Ask the agent to add a Python package to the current project and it will detect the existing dependency manager and install it with the correct command.

Frequently Asked Questions about managing-python-dependencies

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I install a Python package without using global pip?▼

First detect the project's dependency manager from files like uv.lock, pyproject.toml, Pipfile, or requirements.txt, then use that tool's install command. If none exists, create a .venv and install with .venv/bin/pip, never the global pip.

How do I know which Python dependency manager a project uses?▼

Check for signal files in priority order: uv.lock or [tool.uv] in pyproject.toml means uv, [tool.poetry] means Poetry, Pipfile means Pipenv, environment.yml means Conda, and requirements.txt alone means venv plus pip.

Should I use uv or pip for a new Python project?▼

If the project already has a uv.lock or [tool.uv] configuration, use uv add and uv sync. For projects with no dependency manager, the default is venv plus pip with dependencies frozen into requirements.txt.

Can I run Python tests on macOS when the project requires CUDA dependencies?▼

Yes, use the opt-in review environment script setup_review_env.py, which builds a CPU-compatible virtual environment per workspace hash and filters out CUDA and NVIDIA packages. Then run tests via run_in_env.py with pytest or ruff.

Why is running pip install globally a problem?▼

Global pip installs bypass the project's declared dependency manager, causing version conflicts and unreproducible environments. Packages should always be installed through the detected project tooling or an isolated .venv.

Related Skills

Explore More Software Engineering Skills→
juanibiapinajuanibiapina
30

workspace

Organize local GitHub projects by owner/repo in $WORKSPACE.

Community
Intermediate
janewilkinjanewilkin

lint

Lints Python code with ruff and checks types with mypy.

Community
Intermediate
peteromalletpeteromallet
97

llama-cpp

Run CPU-only LLM inference with llama-cpp on non-NVIDIA hardware.

Community
Intermediate
robandersonrobanderson

uv-python-manager

Manage Python projects, dependencies, and virtual environments with UV.

Community
Advanced