azure-cost-optimization
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from Azure APIs, monitoring metrics, and external pricing pages, which could theoretically contain instructions designed to influence the agent's behavior.\n
- Ingestion points:
az restAPI responses,fetch_webpagecontent from Azure pricing domains, and output from theazqrtool.\n - Boundary markers: The skill does not explicitly define delimiters or instructions to ignore embedded instructions when processing these external data sources.\n
- Capability inventory: The skill can write files (
create_file), delete temporary files (Remove-Item), and execute POST requests to the Azure Management API (az rest).\n - Sanitization: No specific sanitization or validation logic is applied to the retrieved content before it is interpolated into the final report.\n- [COMMAND_EXECUTION]: The skill relies extensively on shell commands to interact with the Azure environment.\n
- Evidence: Usage of
az account show,az resource list,az monitor metrics list, andaz restthroughoutSKILL.mdto collect resource and cost data.\n- [EXTERNAL_DOWNLOADS]: The skill requires an external binary dependency to perform its primary function.\n - Evidence:
Azure Quick Review (azqr)is listed as a required tool in Step 0 and is used in Step 2 ofSKILL.mdto identify orphaned resources.
Audit Metadata