frontend-design
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill proactively addresses indirect prompt injection risks associated with runtime data ingestion. In
SKILL.mdandreferences/premium-details.md, it mandates that the agent treat content retrieved via browser-assisted verification (DOM text, console output, screenshots) as untrusted data rather than instructions. This acts as a boundary marker for processing external inputs. - [COMMAND_EXECUTION]: The skill instructions include steps to check
package.jsonand output package installation commands when dependencies are missing. This follows standard development workflows and relies on user review or explicit confirmation for execution. - [DYNAMIC_EXECUTION]: The skill defines strict security boundaries for browser-assisted verification tools (e.g., Playwright, Puppeteer). It requires that JavaScript execution be read-only, restricts access to cookies and credentials, and mandates explicit user confirmation for any mutations like form submissions or API writes.
Audit Metadata