roblox-to-issues
Pass
Audited by Gen Agent Trust Hub on Aug 2, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill defines a project management methodology for breaking down game features into vertical slices. All operations are consistent with its stated purpose.
- [COMMAND_EXECUTION]: The instructions direct the agent to use the GitHub CLI tool (
gh) to publish approved slices as issues. This is a standard and safe use of professional development tooling. - [PROMPT_INJECTION]: The skill ingests external project requirements (PRDs) which could theoretically contain indirect instructions. However, the risk is minimized by the mandatory human review step ("Quiz the user") which requires user approval of all generated tasks before any files are written or issues are created.
- Ingestion points: External PRDs, conversation history, and the
ROADMAP.mdfile. - Boundary markers: Not explicitly defined in the prompt interpolation.
- Capability inventory: File system write access (markdown files) and shell command execution (GitHub CLI).
- Sanitization: Relies on manual user verification and iteration in Step 3 of the process.
Audit Metadata