azure-external-attack-surface-management
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and suggests fetching content from official Microsoft domains (learn.microsoft.com) and the MicrosoftDocs GitHub organization for documentation and tool installation.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a data ingestion surface by instructing the agent to use
fetch_webpageormcp_microsoftdocs:microsoft_docs_fetchto retrieve external documentation. - Ingestion points: Documentation URLs in
SKILL.md(e.g., learn.microsoft.com). - Boundary markers: None explicitly defined in the prompt instructions.
- Capability inventory: Uses
read_filefor local content navigation. - Sanitization: No specific sanitization or filtering of external content is mentioned.
- [SAFE]: No hardcoded credentials, persistence mechanisms, obfuscation, or unauthorized command executions were found. All external resources trace back to official vendor infrastructure.
Audit Metadata