build-agent
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses official orq.ai API and MCP tools for agent orchestration, ensuring all operations remain within the legitimate platform ecosystem.
- [SAFE]: Destructive actions such as deleting agents, knowledge bases, or memory stores are explicitly protected by user confirmation requirements via the AskUserQuestion tool.
- [SAFE]: API documentation and configuration examples follow security best practices by using environment variable placeholders for sensitive credentials instead of hardcoding secrets.
- [SAFE]: The skill provides detailed guidance and templates for implementing constraints and boundaries in system instructions, which mitigates the risk of prompt injection and out-of-bounds behavior in developed agents.
- [SAFE]: Knowledge management and memory configuration instructions emphasize the correct architectural separation of static data (Knowledge Bases) and dynamic user context (Memory Stores), preventing common data leakage or privacy pitfalls.
Audit Metadata