adobe-commerce-ai-layer Rapport d’évaluation.

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
• SKILL.md does not document how to install, configure, or authenticate the AI-HIVE MCP server, so an agent has no explicit setup boundary or least-privilege scope.
• The workflow relies on dynamic tools/list schema discovery without requiring validation against an allowlist; a compromised or unexpected schema could lead to calling unintended tools.
• Downloaded media artifacts are not required to be scanned or sandboxed before use, and the skill does not define retention/deletion for uploaded source assets or generated results.