m365-copilot-extensions
Pass
Audited by Gen Agent Trust Hub on Apr 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides defensive configuration patterns for AI agents. For example, Rule 2 provides grounding instructions that explicitly restrict the agent's knowledge to trusted data sources (SharePoint/Graph), mitigating potential hallucinations or obedience to untrusted training data.
- [SAFE]: Rule 4 promotes secure credential management by advising developers to use OAuth 2.0 instead of hardcoded API keys in manifests.
- [SAFE]: The skill includes guidance on implementing Access Control Lists (ACLs) in Rule 6, ensuring that data ingested via Graph Connectors maintains proper permission boundaries within the Microsoft 365 environment.
Audit Metadata