playful-design-router
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided design requests to determine which sub-skills to invoke. This routing mechanism relies on natural language interpretation, which could be influenced by adversarial input designed to trigger unintended skills.
- Ingestion points: User-provided design requests as described in SKILL.md.
- Boundary markers: None identified.
- Capability inventory: The skill searches project and home directories for installed skills in
.claude/skills/,.agents/skills/, and.cursor/skills/. - Sanitization: None identified.
- [EXTERNAL_DOWNLOADS]: The skill documentation provides commands to install external skills from various repositories, including those by independent developers and Vercel's official GitHub repository. The skill explicitly requires user consent before any installation occurs.
- [COMMAND_EXECUTION]: The documentation suggests using
npx skills addcommands to manage dependencies. These commands involve shell execution that should be monitored for automatic triggering by the agent.
Audit Metadata