cost-estimation
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution via Fallback APIs: The skill provides guidance for using Azure CLI (
az rest) or Azure PowerShell (Invoke-AzRestMethod) as fallback mechanisms if the primary MCP tools are unavailable. This is a standard practice for Azure management tools and includes instructions to avoid using local interpreters like Python or Bash for data parsing, which reduces the risk of unintended code execution. - Data Handling Considerations: The skill manages sensitive information, such as negotiated price sheets and forecast data. It includes explicit guardrails to prevent the exposure of access tokens or pricesheet URLs. Additionally, it specifies that short-lived URLs for pricesheet downloads should not be automatically parsed by the agent, requiring user intervention instead.
- Indirect Prompt Injection Surface: As the skill processes external data from Azure APIs and potentially user-provided price sheet files, it possesses a standard surface for indirect prompt injection. The skill mitigates this by instructing the agent to use tool-side filtering and to avoid local interpretation of results.
Audit Metadata