microsoft-learn-grounding
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill establishes a controlled grounding workflow using official Microsoft Learn tools (
microsoft_docs_searchandmicrosoft_docs_fetch) via an approved API endpoint. - [SAFE]: It includes comprehensive filtering logic in
excluded-sources.mdto prevent the agent from relying on outdated security guidance, archived product versions, or non-authoritative blog posts. - [SAFE]: The instructions incorporate a 'self-reference rule' to ensure that content edits are grounded in independent source material rather than circular references to the article's own published version.
- [SAFE]: The skill provides explicit guidance on detecting feature deprecations within documentation, which improves the reliability and safety of the recommendations provided by the agent.
- [SAFE]: No evidence of prompt injection, data exfiltration, obfuscation, or unauthorized command execution was found in the provided files.
Audit Metadata