pixel-art
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [COMMAND_EXECUTION]: The 'Workflow' section suggests using a 'local preview command' to inspect generated assets. This instruction is open-ended and could lead to the agent executing arbitrary shell commands if not properly constrained by the underlying platform.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input ($ARGUMENTS) to generate SVG content. There is an attack surface where malicious input could be embedded into the SVG code, which is then processed by a preview tool or written to the project files.
- Ingestion points: User instructions provided through the $ARGUMENTS placeholder in SKILL.md.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the prompt template.
- Capability inventory: The skill utilize file writing/editing tools and references the use of a local execution environment for file previews.
- Sanitization: There is no evidence of input validation or escaping before the user data is interpolated into the SVG structure.
- [NO_CODE]: The skill consists entirely of instructional markdown and templates without any bundled executable scripts or code files.
Audit Metadata