code-writing-guide Eval report.

Validation plan

What the agent should do next

1. 1Inspect repository, README/SKILL.md, license, and recent commits before production use.
2. 2Install in an isolated workspace or sandbox with no production secrets available.
3. 3Run the smallest representative task and record files touched, commands run, network access, and outputs.
4. 4Compare the selected skill against at least one alternative when the eval status is review or failed.
5. 5Promote only after the agent reports a successful verification result and unresolved warnings are accepted.

Do not use when

Conditions that require another skill

• teams that need a vendor-supported SLA
• production agents without a repository review
• Low GitHub adoption signal
• The frontmatter uses a YAML block scalar for `description` (`>-`), but the parsed metadata shows the description as `>-`, so skill catalogs using a simple frontmatter parser will display no useful description.
• The skill assumes `unity_play_control`, Rider MCP refactoring tools, and the `edit-scene` skill are available, but it does not list these as prerequisites or define fallback behavior when they are missing.
• Resource paths use `${CLAUDE_SKILL_DIR}` with no mention of how that variable is set or what to do if it is unavailable, which may reduce portability across agent environments.