pipelines-health-check

Pass

Audited by Gen Agent Trust Hub on Mar 20, 2026

Risk Level: SAFE
Full Analysis
  • Standard Tool Integration: The skill utilizes official command-line interfaces (az and gh) to interact with Azure DevOps and GitHub. This is the intended behavior for gathering CI/CD metrics and PR status in a developer environment.
  • Data Sanitization: The PowerShell scripts include a Sanitize-ErrorString function that cleans and truncates external content (such as build logs or error messages) before it is processed. This reduces the risk of malformed data impacting the agent's logic.
  • Trusted External Dependencies: The skill suggests the use of the workiq tool from the @microsoft NPM scope and a specialized binlog analysis tool. These resources originate from trusted organizations and are used to provide additional context for infrastructure failures.
  • Indirect Data Processing: The skill ingests PR titles and build error messages from external repositories. While this is a surface for indirect instructions, the skill implements sanitization and uses structured data templates for its subagent investigations to maintain safety.
  • Least Privilege Auth: The instructions rely on the user's existing local authentication (az login, gh auth login), ensuring the skill operates within the user's established permission boundaries without requiring hardcoded secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 20, 2026, 01:36 AM