container-apps-deployment
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- Command Execution Patterns: The skill includes guidance for executing Azure CLI (
az) and Azure Developer CLI (azd) commands within deployment hooks. It explicitly recommends using argument arrays with functions likeexecFileSyncorspawnSyncinstead of shell-interpolated strings. This approach is a recognized security measure to prevent command injection vulnerabilities. - External Resource References: The documentation references Azure Verified Modules (AVM) from the public Bicep registry (
br/public). These are official infrastructure-as-code components used for standard Azure deployments. Referencing well-known, official registries is common practice for maintaining infrastructure consistency and reliability. - Indirect Prompt Injection Surface: The deployment patterns described involve reading configuration values (such as
API_URL) from theazdenvironment. This creates a data flow from external environment variables into command arguments. To mitigate risks associated with untrusted data, the skill advises against shell parsing and suggests using structured execution methods. - Build Context Sanitization: The skill emphasizes the use of
.dockerignoreto exclude sensitive files like.env, local databases, and Git metadata from being uploaded to Azure Container Registry during the build process. This follows security best practices for protecting local credentials during cloud deployments.
Audit Metadata