azure-horizondb
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill retrieves technical documentation from official Microsoft domains (
learn.microsoft.com). These are recognized as trusted and well-known services for documentation. - [INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface for external content.
- Ingestion points: The
microsoft_docs_fetchandfetch_webpagetools are used to retrieve documentation from the web (found inSKILL.md). - Boundary markers: No specific delimiters or safety warnings for the fetched content are defined in the skill instructions.
- Capability inventory: The skill does not have capabilities for file modification, command execution, or network data transmission.
- Sanitization: No explicit sanitization or verification of the retrieved markdown is performed by the skill.
- [SAFE]: No malicious patterns such as obfuscation, privilege escalation, or persistence were detected. The behavior is consistent with the skill's stated purpose.
Audit Metadata