ink
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user requirements and audit existing icon sets, creating an ingestion surface for potentially untrusted vector data.\n
- Ingestion points: Design briefs, illustration requirements, and existing SVG files for consistency auditing (SKILL.md).\n
- Boundary markers: Relies on platform boundaries and explicit instructions to produce clean code and visually verify optimized output (SKILL.md).\n
- Capability inventory: The skill modifies project documentation and executes asset optimization commands (SKILL.md, reference/svg-optimization.md).\n
- Sanitization: Employs SVGO optimization to strip editor metadata and normalize content, reducing the risk of embedded malicious payloads in vector files (reference/svg-optimization.md).\n- [COMMAND_EXECUTION]: The skill provides procedures for running well-known CLI tools to manage and optimize assets.\n
- Evidence: Includes command recipes for
npx svgoandnpx svgstorefor path optimization and sprite sheet generation (reference/svg-optimization.md).
Audit Metadata