design-orchestration
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a structured workflow for design management without any dangerous command execution, data exfiltration, or obfuscation.
- [PROMPT_INJECTION]: No evidence of direct malicious instructions or attempts to override agent safety filters. The skill ingests user-provided feature proposals as defined in SKILL.md which presents a surface for indirect prompt injection. No boundary markers or sanitization protocols are explicitly defined for this input. However, the skill's capability inventory is limited to routing and invoking other brainstorming tasks, with no access to sensitive files or network tools.
- [COMMAND_EXECUTION]: There are no shell commands or subprocess calls present in the skill instructions.
- [DATA_EXFILTRATION]: No network access or data exfiltration patterns were detected.
Audit Metadata