low-code-generation
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted natural language descriptions and schema definitions to generate UI code, creating a surface where malicious instructions could be embedded in the data.
- Ingestion points: User-provided schemas and descriptions processed through the logic defined in SKILL.md.
- Boundary markers: The skill does not employ explicit delimiters or instructions to the agent to disregard embedded instructions within the input schemas.
- Capability inventory: The skill generates React components and TypeScript logic, allowing data to be transformed into executable structures.
- Sanitization: The implementation uses JSON.stringify for conditional logic values, but other properties like field labels and regex patterns are interpolated directly into code templates.
- [DYNAMIC_EXECUTION]: The skill performs runtime generation of executable scripts and components using string interpolation of schema properties.
- Evidence: The generateForm and generateTable functions in SKILL.md demonstrate the assembly of JavaScript/React code snippets from dynamic input variables.
Audit Metadata