dt-obs-azure

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE]: The skill contains no executable scripts or binaries. It is composed of Markdown files with DQL query templates and documentation.
  • [SAFE]: The skill performs read-only metadata queries within the Dynatrace environment. No evidence of data exfiltration, credential theft, or unauthorized command execution was found.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a surface area for indirect prompt injection via the ingestion of Azure resource metadata.
  • Ingestion points: Azure resource names, tags, and JSON objects (azure.object) retrieved via DQL.
  • Boundary markers: Absent. The templates do not use delimiters for untrusted data.
  • Capability inventory: None. The skill does not contain subprocess calls, file-write, or network operations.
  • Sanitization: Absent. Cloud metadata is processed directly.
  • Conclusion: The risk is negligible due to the lack of exploitable capabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 03:41 PM