icon-set-generator
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input to generate visual assets and a preview HTML page, creating a surface for indirect prompt injection.
- Ingestion points: User-provided project descriptions, business names, and icon names are ingested in SKILL.md (Step 1).
- Boundary markers: The instructions do not specify any boundary markers or instructions for the agent to ignore potentially malicious content embedded in the user's project description.
- Capability inventory: The skill uses file-writing capabilities to generate .svg, .json, and .html files in the icons/ directory (SKILL.md, Step 4 and 5).
- Sanitization: While SKILL.md includes SVG Rules that limit elements (Rule 8) and attributes (Rule 5), there are no explicit sanitization steps for user-provided strings (like the project name) before they are interpolated into the preview.html template (references/preview-template.md).
Audit Metadata