azure-ai-vision
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches content from official Microsoft documentation at
learn.microsoft.comand references the officialMicrosoftDocsGitHub organization for tool installation. These are trusted, well-known services. - [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest external documentation content via
microsoft_docs_fetchorfetch_webpage. This represents an indirect prompt injection surface where instructions could be hidden in external data. - Ingestion points: External documentation fetched from
learn.microsoft.com. - Boundary markers: None specified in the instructions.
- Capability inventory: Web fetching tools (
microsoft_docs_fetch,fetch_webpage). - Sanitization: None specified in the instructions.
Audit Metadata