ux-review
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from external URLs and screenshots for its audit process, which introduces a risk of indirect prompt injection if the audited site contains malicious instructions.
- Ingestion points: Page content fetching and screenshot capture described in references/audit.md.
- Boundary markers: No explicit delimiters or instructions are provided to the agent to treat website content as untrusted data rather than instructions.
- Capability inventory: The agent has capabilities for command execution via npx and file system writes to the .jimmy/work/ directory.
- Sanitization: There is no requirement for sanitization or filtering of the fetched web content.
- [COMMAND_EXECUTION]: The skill instructions suggest that the agent should execute shell commands, specifically npx playwright, to capture visual evidence of design systems.
- [EXTERNAL_DOWNLOADS]: The recommendation to use npx involves the dynamic download of packages from the npm registry. While Playwright is a well-known tool from a reputable source, it represents an external code dependency.
Audit Metadata