guided-gui-design
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The primary content of the skill consists of documentation, checklists, and theoretical foundations for GUI design. No malicious code, prompt injections, or obfuscation techniques were identified across the 18 analyzed files.
- [EXTERNAL_DOWNLOADS]: The project uses the
compose-agentsmdtool to synchronize global agent rules from a vendor-controlled repository (github:metyatech/agent-rules@HEAD). This is a legitimate configuration management practice within the author's ecosystem. - [COMMAND_EXECUTION]: The skill includes a global ruleset (
AGENTS.md) that defines procedures for shell command execution, privilege management (sudo usage), and developer workflow automation. These rules are provided to ensure consistent and safe agent operations rather than to execute harmful commands. - [PROMPT_INJECTION]: The instructions in
SKILL.mduse standard compliance vocabulary (e.g., "The agent MUST NOT...") to direct agent behavior for GUI reviews and do not include patterns intended to bypass safety filters or override system constraints.
Audit Metadata