azure-data-factory

Pass

Audited by Gen Agent Trust Hub on Sep 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides numerous links to official Azure documentation at learn.microsoft.com. It instructs the agent to retrieve this content dynamically using the mcp_microsoftdocs:microsoft_docs_fetch or fetch_webpage tools. These references target a well-known service and are used as intended for technical knowledge retrieval.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external markdown data from documentation pages. Although the source is trusted, the skill's architecture involves processing untrusted external text which is a theoretical surface for indirect prompt injection; however, no exploitable capabilities (like file writes or command execution) are present in the skill to leverage such an injection.
  • [SAFE]: The skill serves as a knowledge index. It does not include any scripts, hardcoded secrets, obfuscation, or persistence mechanisms. The suggested installation link for the mcp_microsoftdocs tool points to an official repository under the MicrosoftDocs GitHub organization.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 29, 2026, 09:49 AM