review-web-design
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and analyze external content from URLs using the WebFetch tool. This introduces a surface for indirect prompt injection where an attacker-controlled webpage could contain instructions meant to hijack the agent's behavior.
- Ingestion points: External website content and source code fetched via the WebFetch tool during the design review process (Step 1, Step 3, Step 5).
- Boundary markers: The instructions do not include specific delimiters or warnings to the agent to treat fetched content strictly as data rather than instructions.
- Capability inventory: The skill allows the use of file system tools (Read, Grep, Glob) alongside network capabilities (WebFetch), which could be chained if a malicious prompt is successfully injected from a remote site.
- Sanitization: There is no explicit requirement for the agent to sanitize or filter the content retrieved from external sources before processing it.
Audit Metadata