skills/microsoft/waza/azd-publish/Gen Agent Trust Hub

azd-publish

Pass

Audited by Gen Agent Trust Hub on Jun 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • Command Execution: The skill employs standard development utilities such as git and the GitHub CLI (gh) to manage branches, commits, and pull requests. These operations are essential for the release workflow and are designed to prompt for user confirmation at decision points.
  • Indirect Prompt Injection Surface: The skill ingests data from git log to summarize changes for the changelog and pull request body. Processing external text such as commit messages introduces a surface where unexpected content could influence the agent's summarized output. The skill addresses this by requesting user review of the summarized changes before they are finalized.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 27, 2026, 04:51 PM
Security Audit — agent-trust-hub — azd-publish