azure-functions-deploy

Pass

Audited by Gen Agent Trust Hub on Oct 23, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection Surface]: The skill analyzes local project files, such as azure.yaml and host.json, to determine the appropriate deployment workflow. While this is necessary for project identification, it presents a potential surface where malformed configuration files could influence agent behavior.\n
  • Ingestion points: Project configuration files including azure.yaml, host.json, and infrastructure definitions in infra/.\n
  • Capability inventory: Access to Azure CLI (az), Azure Developer CLI (azd), and Node package execution (npx).\n
  • Sanitization: The skill relies on specific field matching and file existence checks rather than broad content processing.\n- [External Package Execution]: After a successful deployment, the skill is instructed to run a telemetry collector using npx -y @azure/functions-skills@latest. This involves downloading and executing a package from a remote registry. As this is a resource managed by a recognized vendor for this skill, it is generally considered part of the tool's standard operation.\n- [Administrative Command Execution]: The skill provides commands for the agent to monitor deployment progress via az group show and az deployment group list. These are standard use cases for the Azure CLI within a deployment context.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 23, 2026, 08:11 AM