spec-brainstorm
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for conversational requirements gathering and architecture design. It guides the user through a structured process to produce markdown specification files in a dedicated directory.
- [DATA_EXPOSURE]: The skill instructions involve reading project-specific files such as README, AGENTS.md, and source code implementation. This access is localized to the project context to inform the design process and does not target sensitive system files or credentials.
- [COMMAND_EXECUTION]: The skill mentions using an internal tool, 'oracle-architect', for modeling tasks. There are no patterns of arbitrary shell execution, privilege escalation, or persistence mechanisms detected.
Audit Metadata