azure-cloud-hsm
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches documentation from official Microsoft domains (learn.microsoft.com) and references the vendor's GitHub repository for tool setup.
- [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: Retrieves external content via mcp_microsoftdocs:microsoft_docs_fetch and fetch_webpage tools from official Azure documentation URLs specified in the category tables.
- Boundary markers: Absent in the skill instructions; instructions rely on the model's ability to distinguish fetched content from system instructions.
- Capability inventory: The skill is restricted to documentation retrieval and guidance; no capabilities for file writing, system command execution, or data exfiltration were found.
- Sanitization: Documentation is fetched from a trusted vendor source in Markdown format.
Audit Metadata