# 064-easy_d43ff184

> Records common pitfalls and assumptions for Python module execution and GitHub Actions sparse checkout setups.

- Skill: `tools-only/064-easy-d43ff184` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds add tools-only/064-easy-d43ff184`
- Raw SKILL.md: https://api.skillmd.com/api/skills/tools-only/064-easy-d43ff184/raw
- Safety review: PASS (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools, Dev Tooling
- Tags: Cli Entrypoint, Github Actions, Pyproject Toml, Python, Sparse Checkout
- Author: tools-only (https://skillmd.com/u/tools-only)
- Updated: 2026-08-22
- Page: https://skillmd.com/skills/tools-only/064-easy-d43ff184

---

# Don't Forget

Some important but easy-to-forget assumptions and gotchas that can cause headaches:

- `python -m` uses dot notation (for example, `python -m scripts.resources.parse_issue_form`), not slash paths or `.py` suffixes.
- `python -m` only works for modules with a CLI entrypoint (`if __name__ == "__main__":`).
- GitHub Actions with sparse checkout must include `pyproject.toml` so `find_repo_root()` can locate the repo root.
- If a workflow runs scripts that import `scripts.*`, either run from the repo root or set `PYTHONPATH` to the repo root.
- Sparse checkout must include any data files the script reads (for example, `THE_RESOURCES_TABLE.csv`, `templates/`).
- GitHub Actions using sparse checkout must include `pyproject.toml` so scripts can locate the repo root (via `find_repo_root()`).