azure-defender-for-cloud
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with remote content from learn.microsoft.com, which serves as an ingestion point for external data into the agent's context. \n
- Ingestion points: Documentation retrieval from learn.microsoft.com using fetch tools specified in SKILL.md. \n
- Boundary markers: The instructions do not define delimiters to encapsulate or isolate the fetched documentation from other system instructions. \n
- Capability inventory: No scripts or tools with dangerous capabilities (e.g., file writing, remote code execution) are present in the skill. \n
- Sanitization: No specific mechanisms for sanitizing or filtering retrieved content are provided. \n
- Note: The ingestion is limited to trusted vendor documentation, significantly reducing the risk of malicious indirect instruction. \n- [EXTERNAL_DOWNLOADS]: The skill contains references to documentation and tools hosted by Microsoft and GitHub (github.com/MicrosoftDocs). These are official vendor resources and are considered safe within the scope of this analysis.
Audit Metadata