using-agent-skills 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 appears truncated at the end ('Definition of Done: tests pass, no regres...') and does not include the full operational guidance.
• Permission surface may require sandboxing
• The meta-skill references many other skills but does not explain how to load or invoke a chosen skill, nor what to do when no skill matches.
• No explicit limitations, security considerations, or safe operating boundaries are documented for the meta-skill itself.