eval-suite-planner
Pass
Audited by Gen Agent Trust Hub on Apr 9, 2026
Risk Level: SAFE
Full Analysis
- [Trusted Source Integration]: The skill integrates knowledge from official Microsoft documentation and GitHub repositories. These references are used to ground the evaluation guidance in established best practices for agent development and quality assurance.
- [Functional Tool Integration]: The skill is designed to output its findings into document and spreadsheet formats. This utilizes the environment's standard file generation capabilities to provide users with portable and editable evaluation plans.
- [Proactive Safety Planning]: The instructions mandate the inclusion of adversarial and safety-related scenarios in every generated plan. This encourages developers to consider security and robustness as core components of their agent's lifecycle.
- [Input Processing]: The skill processes descriptive input about an agent to categorize and plan tests. It does not perform network operations on untrusted domains or execute dynamic code based on the user's input, maintaining a secure operational boundary.
Audit Metadata