AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Python

skill-narumiruna-skills-python · by narumiruna

Use when a task involves Python project setup or standalone scripts with uv, including dependency management, `uv run`, `uv run --with`, `--no-project`, inline script metadata, quality gates (ruff, ty, pytest, coverage, prek/pre-commit), and package build or publishing workflows.

No reviews yet
0 installs
43 views
0.0% view→install

Install

$ agentstack add skill-narumiruna-skills-python

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-narumiruna-skills-python)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Python? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Python

Core Posture

Use uv as the default workflow for Python projects, standalone scripts, dependency management, quality checks, and package release. Prefer the repository's existing configuration and documented commands over introducing new tooling.

Mode Selection

  • Choose project mode when the work has a pyproject.toml, shared package code, or lockfile.
  • Treat a brand-new repository created with uv init as project mode.
  • Choose standalone-script mode when the work is a single file, stdin snippet, or ad hoc invocation that should not become a full project.
  • Treat a script inside a project as project mode when it imports local package code.
  • Use --no-project only when a script must ignore the surrounding project.

Project Workflow

  1. Inspect pyproject.toml, lockfiles, repository docs, and existing test/tooling choices before changing dependencies or commands.
  2. For a brand-new project, run uv init , then install the default dev toolchain with uv add --dev ruff ty pytest pytest-cov.
  3. Use pytest as the default test framework for new projects: function-based tests/test_*.py files with plain assert.
  4. Add dependencies with uv add; remove them with uv remove; reconcile external dependency changes with uv sync.
  5. Run project Python, tests, and tools through uv run unless the repository provides a documented wrapper command; for pytest, default to an explicit tests target when the project does not document another test location.
  6. Run the repository quality gate before finalizing dependency or code changes.

Standalone Script Workflow

  1. Decide whether the script should use project dependencies, one-off --with dependencies, or inline metadata.
  2. Use plain uv run script.py when no extra dependencies are needed.
  3. Use uv run --with ... script.py for disposable per-invocation dependencies.
  4. Use uv init --script and uv add --script when dependencies should live with the script.
  5. Put --no-project before the script name, and never use it for a script that imports local package code.
  6. Use uv run --python or requires-python in metadata when the script needs a specific Python version.
  7. Read references/scripts.md for stdin, heredoc, shebang, alternate indexes, Windows .pyw, locking, and reproducibility details.

Non-Negotiable Rules

  • Manage dependencies with uv add, uv remove, and uv sync; do not use pip install to mutate a project environment.
  • Run Python, pytest, scripts, and Python tools with uv run unless a repository command wraps them; when no project-specific pytest target is documented, include tests explicitly (for example, uv run pytest tests) rather than bare uv run pytest.
  • Follow existing repository choices; do not migrate test frameworks, hook runners, or tool configuration unless asked.
  • For a brand-new project, immediately add ruff, ty, pytest, and pytest-cov; do not start with unittest, unittest.TestCase, or class-based Test* suites.
  • Do not create a pyproject.toml only to run a one-file script.
  • Treat uv run --with as disposable; use inline metadata when the script should be shared or reused.
  • Build release artifacts with uv build --no-sources so local [tool.uv.sources] overrides do not leak into release output.
  • Test installs from the built wheel or published artifact, not only from the source checkout.

Quality Gate Priority

Use this order before finalizing code, dependency, or packaging changes:

  1. The repository's documented aggregate gate.
  2. prek run -a when the repo uses prek.
  3. The documented pre-commit command when the repo uses pre-commit and provides no prek path.
  4. Individual uv run ... commands only when no aggregate gate exists or when narrowing failures.

Do not introduce or switch hook runners only for verification. Read references/quality.md for command details, coverage, CI examples, and fallback checks.

Release Workflow

  1. Run the repository quality gate.
  2. Build artifacts in dist/ with uv build --no-sources.
  3. Inspect wheel and sdist contents.
  4. Test install from the built wheel or published artifact in a fresh uv-managed invocation.
  5. Read references/packaging.md for publish commands, Test PyPI flow, and artifact inspection details.

Common Mistakes

  • Treating a one-file script as project work when inline metadata or --no-project would be smaller and clearer.
  • Using --no-project for a script that imports local package code.
  • Keeping reusable script dependencies only in uv run --with ... shell history instead of the script metadata.
  • Starting a brand-new project with unittest or without the default dev toolchain.
  • Running tools outside uv.
  • Publishing before verifying artifacts.

References

  • references/quality.md - Full commands, CI examples, coverage, prek, and pre-commit fallback usage.
  • references/packaging.md - Build, inspect, and publish details.
  • references/scripts.md - Standalone script patterns, inline metadata, locking, and special cases.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.