react
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent on how to handle and review user-supplied React code and dynamic templates. It includes specific defensive measures to ensure that instructions hidden within data (such as HTML comments or template variables) are not misinterpreted as commands.
- Ingestion points: Processes and reviews user-supplied React components and template data described in SKILL.md.
- Boundary markers: The instructions explicitly require the use of escaping and validation to create boundaries between data and executable code.
- Capability inventory: The skill is strictly instructional and does not invoke subprocesses, network operations, or sensitive file system access.
- Sanitization: Mandates the use of strict allow-lists, output escaping, and sanitization for all dynamic content and user-provided variables.
Audit Metadata