forge
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill demonstrates high security awareness by incorporating detailed warnings and defensive strategies against common AI-related attack vectors, such as slopsquatting (maliciously registered phantom packages) and prompt injection.
- [SAFE]: External dependencies, including CDN-hosted libraries like D3 from unpkg.com, are handled with best-practice requirements for version pinning and Subresource Integrity (SRI) hashes to prevent supply-chain attacks.
- [SAFE]: The skill provides robust mitigation strategies for indirect prompt injection, specifically recommending the use of sanitizers like DOMPurify and rehype-sanitize for any AI-generated content rendered in a browser.
- [SAFE]: The handoff process (e.g., Forge-to-Builder) includes mandatory security review steps and technical debt recording, ensuring that prototype shortcuts are explicitly identified and addressed before production integration.
Audit Metadata