design-engineer-router
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill acts as a logic-based router for design tasks.
- [PROMPT_INJECTION]: The instructions include directives to trigger on every design-related task. This is consistent with the skill's purpose as an orchestrator and does not involve attempts to override AI safety guardrails or extract sensitive system information.
- [DATA_EXPOSURE]: The skill references standard local paths where other skills might be installed (e.g.,
~/.cursor/skills/). It does not attempt to read sensitive files, credentials, or environment variables. - [EXTERNAL_DOWNLOADS]: No remote scripts or packages are downloaded or executed. The skill references external individuals and their websites (
emilkowal.ski,jakub.kr) as role models for design standards, but these are plain-text informational links. - [INDIRECT_PROMPT_INJECTION]: The skill operates on user-provided design tasks. While it instructs the agent to load and read other local skill files (
SKILL.md), it does not ingest untrusted remote data that could lead to indirect injection within this specific routing logic.
Audit Metadata