azure-ai-contentunderstanding-py
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the official Azure AI Content Understanding library (
azure-ai-contentunderstanding) and the Azure Identity library from public registries. These are official packages from a well-known technology provider. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external URLs (documents, images, audio, video) which is a standard surface for indirect prompt injection.
- Ingestion points: The
AnalyzeInputclass accepts remote URLs in multiple functions acrossSKILL.md(e.g.,client.begin_analyze). - Boundary markers: There are no instructions or delimiters defined to separate extracted content from agent instructions.
- Capability inventory: The skill has the capability to perform network requests to Azure services and read file contents.
- Sanitization: There is no evidence of sanitization or filtering applied to the extracted markdown or transcript data before it is presented to the agent context.
Audit Metadata