web-design-reviewer
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it processes untrusted data from external URLs.
- Ingestion points: The skill navigates to user-provided URLs and retrieves DOM snapshots and visual data in
SKILL.md(Step 2.1). - Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore or isolate embedded commands within the fetched web content.
- Capability inventory: The skill has the ability to search the local workspace using
grep_searchandsemantic_search, and it can perform direct source code modifications across various frameworks (React, Next.js, etc.) inSKILL.md(Step 3). - Sanitization: No sanitization or filtering of the external DOM content is mentioned before it is processed by the agent to determine code fixes.
- [COMMAND_EXECUTION]: The skill uses automated search tools and code editing capabilities to apply design fixes. These operations are driven by the analysis of external web content, which could be manipulated by an attacker to trigger unintended file system operations or malicious code injections into the local project.
Audit Metadata