fix(cli): add allow_unicode=True and encoding="utf-8" to YAML I/O

[seiya-koji]

Description

None of the yaml.dump() calls specify allow_unicode=True,
causing non-ASCII characters in extension descriptions to be escaped to \uXXXX sequences
in generated .agent.md frontmatter and config files.

Additionally, most write_text() and read_text() calls for YAML config files lack
encoding="utf-8", making them locale-dependent.
On Windows (where the default encoding is cp932 or cp1252),
this can cause UnicodeEncodeError on write or UnicodeDecodeError on read
when non-ASCII characters are present.

# Before (escaped)
description: "Pr\u00FCfe Konformit\u00E4t der Implementierung"

# After (fixed)
description: "Prüfe Konformität der Implementierung"

Reproduction

1. Create an extension with non-ASCII description in extension.yml
2. Run specify extension add --dev <path>
3. Inspect generated .agent.md — frontmatter contains \uXXXX escapes

Fix

Add allow_unicode=True to all yaml.dump() calls, and encoding="utf-8" to all
corresponding write_text() / read_text() calls.

Testing

• Tested locally with uv run specify --help
• Ran existing tests with uv sync && uv run pytest
• Tested with a sample project (if applicable)

AI Disclosure

• I did not use AI assistance for this contribution
• I did use AI assistance (describe below)

The bug was discovered when installing a Japanese-language extension via specify extension add
through GitHub Copilot (Claude Opus 4.6).
Root cause analysis and fix were also generated by the same agent.

[Copilot]

Pull request overview

This PR addresses Unicode handling in YAML serialization/deserialization across the CLI so non-ASCII characters are preserved (not \uXXXX-escaped) and YAML config reads/writes are no longer locale-dependent (notably fixing Windows default-encoding pitfalls).

Changes:

• Add allow_unicode=True to relevant yaml.dump() calls so YAML output preserves non-ASCII characters.
• Add encoding="utf-8" to YAML-related read_text() / write_text() calls.
• Add a unit test to ensure rendered frontmatter retains Unicode characters.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
tests/test_extensions.py	Adds regression coverage ensuring YAML frontmatter rendering doesn’t escape Unicode.
src/specify_cli/presets.py	Forces UTF-8 when reading preset catalog YAML config.
src/specify_cli/extensions.py	Forces UTF-8 for extension YAML configs and enables Unicode-preserving YAML dumps for saved config.
src/specify_cli/agents.py	Enables Unicode-preserving YAML rendering for command frontmatter.
src/specify_cli/__init__.py	Forces UTF-8 read/write and enables Unicode-preserving dumps for catalog-related CLI config updates.

Comments suppressed due to low confidence (1)

src/specify_cli/extensions.py:982

• With encoding="utf-8", config_path.read_text(...) may raise UnicodeDecodeError for non-UTF-8 files. This isn’t currently caught, so the intended ValidationError wrapping won’t happen and a raw exception may bubble up. Consider catching UnicodeError (or UnicodeDecodeError) alongside yaml.YAMLError/OSError.

            data = yaml.safe_load(config_path.read_text(encoding="utf-8")) or {}
        except (yaml.YAMLError, OSError) as e:
            raise ValidationError(
                f"Failed to read catalog config {config_path}: {e}"
            )

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 5 out of 5 changed files in this pull request and generated no new comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[seiya-koji]

Apologies — my coding agent ran force push without my approval. Sorry for the messy history.

[Copilot]

Pull request overview

Copilot reviewed 5 out of 5 changed files in this pull request and generated no new comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.