azure-cognitive-search

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches documentation from the official Microsoft Learn domain (learn.microsoft.com) and references the vendor's own repository on GitHub (github.com/MicrosoftDocs) for tool installation.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external content retrieved from the web.
  • Ingestion points: Remote documentation content fetched via mcp_microsoftdocs:microsoft_docs_fetch or fetch_webpage in SKILL.md.
  • Boundary markers: None explicitly defined in the skill instructions to separate external data from system instructions.
  • Capability inventory: The skill utilizes network access to retrieve information for the agent's context.
  • Sanitization: No explicit validation or escaping of the fetched documentation is described, though it targets exclusively trusted vendor domains.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 03:20 PM
Security Audit — agent-trust-hub — azure-cognitive-search