antislop-web
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted design briefs and client requirements to generate project configuration files (DESIGN.md), component code, and package installation commands. This creates a surface where a malicious project brief could attempt to override the skill's constraints or manipulate generated code.
- Ingestion points: Processes project-specific briefs and client-provided brand data (e.g., palette, typography) during the generation of the
DESIGN.mdfile. - Boundary markers: The skill lacks explicit instructions or delimiters to distinguish between the developer's instructions and untrusted content from the project brief.
- Capability inventory: The skill is capable of writing files to the project root (
DESIGN.md), generating UI component code (React/TypeScript/Tailwind), and instructing the agent to run package installation commands (e.g.,npm install). - Sanitization: No sanitization or validation logic is defined for the content ingested from the design brief.
Audit Metadata