azure-diagnostics
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from Azure logs and application telemetry, creating a potential surface for indirect prompt injection.
- Ingestion points: Untrusted data is ingested through
az containerapp logsand KQL queries (e.g.,AppExceptions,AppRequests) inreferences/kql-queries.md. - Boundary markers: The skill does not define specific markers or instructions to isolate log content from other prompt instructions.
- Capability inventory: The skill is limited to diagnostic read operations using CLI tools and MCP commands; it lacks write permissions, file system modification capabilities, or network exfiltration tools.
- Sanitization: There is no evidence of log data sanitization or filtering before it is processed by the agent.
- [SAFE]: The skill relies on well-known Azure diagnostic tools and provides references to official Microsoft documentation (
learn.microsoft.com), demonstrating a legitimate and safe functional scope.
Audit Metadata