design-template-monochrome
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts on existing codebases, creating a surface area where data within user files could theoretically influence agent behavior.
- Ingestion points: The agent reads project architecture and tech stacks using the Read, Glob, and Grep tools.
- Boundary markers: No specific instructions are provided to delineate untrusted data from the agent's core instructions.
- Capability inventory: The skill leverages Write and Edit tools to implement design changes on the local file system.
- Sanitization: There is no explicit logic defined for sanitizing or escaping content read from external files.
Audit Metadata