fix(skills): decode skill bodies as UTF-8 🤖🤖🤖

[Sampoornnagpal]

What does this PR do?

Skill metadata is read as UTF-8, but _parse_skill_md() reads the body using the locale encoding. A UTF-8 skill containing 50€ → café 中文 🚀 becomes corrupted instructions on a cp1252 host. Read the body as UTF-8 too.

Add a regression that forces a legacy default even on UTF-8 hosts and checks the loaded skill's full body. The test fails before the fix.

Related issues

The parser remains in use in #161; this change is limited to its body decoding and does not alter that refactor's APIs or arbitrary-file encoding policy.

Validation

uv run pytest -q tests/unit/test_skill.py -k 'not run_script' — 19 passed, 7 Bash-dependent tests deselected.

Windows/Python 3.12 with external import-only helpers for #84/#85 (fcntl/SIGUSR2). Helpers are not included; POSIX locks/signals and Bash execution are not validated by this run.

Checklist

• Ruff lint and formatting pass.
• Regression and skill-loading tests pass.
• Existing UTF-8 skill format retained.
• Existing SPDX headers retained.

Summary by CodeRabbit

• Bug Fixes

  • Skill files are now consistently read as UTF-8, preserving Unicode characters regardless of the platform’s default text encoding.
  • Skill descriptions continue to be parsed correctly when files contain Unicode content.

• Tests

  • Added regression coverage for loading path-based skills with Unicode content under differing default encodings.
  • Added explanatory documentation to related test cases.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 7e79754a-7dee-43c5-95ba-fc86e59924b8

📥 Commits

Reviewing files that changed from the base of the PR and between 5d22bf3 and fd31243.

📒 Files selected for processing (1)

• tests/unit/test_skill.py

🚧 Files skipped from review as they are similar to previous changes (1)

• tests/unit/test_skill.py

Included review availability: Your plan provides up to 12 included reviews per hour; 5 remain after this review.

---

📝 Walkthrough

Walkthrough

The skill loader now reads SKILL.md files as UTF-8. Tests verify that multilingual text, symbols, and emoji remain intact when the platform default encoding is cp1252.

Changes

Skill file encoding

Layer / File(s)	Summary
Explicit UTF-8 skill loading src/nooa/skill.py, tests/unit/test_skill.py	_parse_skill_md explicitly decodes SKILL.md as UTF-8. Regression coverage verifies Unicode preservation and documents related skill construction tests.

Estimated code review effort: 1 (Trivial) | ~5 minutes

Merge Risk: ⚪ Minimal · up to fd312

Skill bodies containing multilingual text and symbols now load as UTF-8 rather than depending on the host locale. The targeted regression coverage indicates no remaining merge-blocking risk.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly identifies the main change: decoding skill bodies as UTF-8. The emojis add minor noise, but they do not obscure the change.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 4 functions across 2 files.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands.}

[alessiodevoto]

LGTM! Thanks for this fix @Sampoornnagpal :)