design-canvas
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes shell commands (
ls,cat, andgrep) to discover brand assets and parse design system configurations, such astailwind.config.*and.cursor/rules/*.md. These commands are scoped to discovery within the project directory. - [INDIRECT_PROMPT_INJECTION]: The skill processes contents from external project files to inform its design philosophy, creating a vulnerability surface where malicious instructions in those files could influence agent behavior.
- Ingestion points: Local project assets and configuration files, including
tailwind.config.js,docs/THEME_GUIDE.md, and rule files in.cursor/rules/. - Boundary markers: None. The skill does not employ specific delimiters or system instructions to ignore potential commands embedded within the discovered design assets or documentation.
- Capability inventory: The skill possesses file system discovery capabilities and the ability to write new files in
.md,.pdf, and.pngformats. - Sanitization: The skill lacks sanitization or validation mechanisms for the data ingested from the project environment.
Audit Metadata