deploy

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection Surface]: The skill reads project configuration from memory-bank.md, which is a surface where data could potentially influence agent behavior.
  • Ingestion points: Project name and environment are read from memory-bank.md in the project root.
  • Boundary markers: No explicit delimiters are used to separate configuration data from instruction context.
  • Capability inventory: The skill can execute shell commands and has write access to the filesystem.
  • Sanitization: Configuration content is used directly in the workflow without an explicit validation or escaping layer.
  • [Command Execution]: The skill executes build and deployment commands through npm and the Power Platform CLI (pac). These are standard operations for development workflows.
  • [External Package Dependency]: The skill may install the @microsoft/power-apps-cli package globally when running on macOS to bypass known authentication issues. This is an official vendor package used for its intended purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 11:50 PM
Security Audit — agent-trust-hub — deploy