azure-synapse-analytics

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches documentation from official Microsoft Learn URLs. This utilizes a well-known and trusted service for technical information, and the scope of external access is limited to these authoritative domains.\n- [SAFE]: Proactive security measures are integrated into the prompt instructions. The skill explicitly requires the agent to treat fetched text as untrusted reference data, ignore any embedded instructions or tool requests, and independently validate any commands before execution, which is a best-practice defense against indirect prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 05:50 PM
Security Audit — agent-trust-hub — azure-synapse-analytics