ai-agent-development
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a production checklist and automated scripts (
validate-agent-checklist.ps1) to ensure agents are built with security in mind, specifically checking for hardcoded credentials and missing error handling. - [EXTERNAL_DOWNLOADS]: Recommends the installation of standard industry libraries from Microsoft and OpenTelemetry, such as
agent-framework-azure-ai,azure-ai-projects, andopentelemetry-apivia standard package managers. - [COMMAND_EXECUTION]: Includes utility scripts like
scaffold-agent.pywhich are designed to create local project directories and boilerplate files, facilitating standard development workflows. - [PROMPT_INJECTION]: The skill includes guidance on mitigating prompt injection and references the OWASP Top 10 for AI Security to educate developers on common attack vectors.
- [SAFE]: Implements best practices for credential management by advocating for the use of environment variables and Azure Key Vault instead of hardcoding sensitive information.
Audit Metadata