guardrails-reviewer 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
• SKILL.md references `references/language-policy.md` and `references/workflow-hooks.md`, but those files are not present in the submitted skill directory, which may make some instructions incomplete.
• No OpenAgentSkill engagement data yet
• No explicit security guidance for handling potentially untrusted Guardrails documents; the skill should treat incoming documents as data and avoid executing any commands or following instructions embedded in them.
• The skill is mostly written in Chinese, while the agent interface is in English; this is acceptable due to the language policy, but could be clearer for non-Chinese users if language-policy.md is not available.