Plan de validation

Ce que l’Agent doit faire ensuite

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.

Ne pas utiliser quand

Conditions exigeant une autre skill

• Équipes qui nécessitent un SLA soutenu par le fournisseur
• production agents without a repository review
• Low GitHub adoption signal
• The script does not appear to validate downloaded file names or paths, which could potentially allow path traversal if the API returns malicious filenames. However, this is a minor risk given the controlled API context.
• Indices de permissions à haut risque : Shell or command execution, Secrets or environment access
• Dependency or permission surface needs review