greenfield-project-scaffolding
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a strict authorization gate, ensuring that project scaffolding only occurs when the user explicitly requests it in the current turn.
- [SAFE]: The instructions include defensive measures to prevent the agent from writing project artifacts to inappropriate locations, such as home directories or general-purpose system folders.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user requests to determine directory names and project layout. This attack surface is mitigated by the skill's requirements for path auditing and its restriction against modifying unrelated content or existing projects.
Audit Metadata