update-avm-modules-in-bicep

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches module version information and documentation from official sources, including the Microsoft Container Registry (mcr.microsoft.com) and the Azure Bicep Registry GitHub repository (github.com/Azure).
  • [COMMAND_EXECUTION]: The skill uses standard Bicep tooling (bicep lint and bicep build) to validate that module updates maintain file integrity and compliance.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes external Bicep files and has the capability to execute shell commands and modify files.
  • Ingestion points: The target Bicep file (${file}) is scanned for module definitions.
  • Boundary markers: The skill does not explicitly define delimiters to isolate instructions from the data within the processed Bicep file.
  • Capability inventory: Includes file writing (#editFiles), network requests (#fetch), and command execution (#runCommands).
  • Sanitization: No explicit sanitization or filtering of the input Bicep file content is specified prior to processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 06:51 AM
Security Audit — agent-trust-hub — update-avm-modules-in-bicep