hiring-intelligence 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
• SKILL.md does not explicitly warn agents that external content from job posts or company profiles could contain prompt injection attempts and must be treated as untrusted data.
• Financial research output is not financial advice; require human review before any live investment decision
• The skill relies entirely on sandbase_describe_tool for parameter schemas, so it is less self-contained if the SandBase gateway is unavailable or the tool definitions change.
• No fallback or error-handling guidance is provided for cases where a selected tool is down, returns malformed data, or lacks permission.