html-prototype
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the
pdf-parsepackage from the npm registry to facilitate text extraction from PDF files during the prototype generation process. - [EXTERNAL_DOWNLOADS]: The skill's documentation suggests the installation and use of well-known third-party tools and services, including Cloudflare Tunnel (
cloudflared), Surge.sh, and Vercel, for the purpose of hosting and sharing the generated HTML prototypes. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it analyzes the content of untrusted documents to determine the layout and logic of the code it generates.
- Ingestion points: The skill reads
.pdf,.txt, and.mdfiles provided by the user or found in thescreen-design/workspace folder. - Boundary markers: No specific delimiters or safety instructions are implemented to prevent instructions embedded within those documents from influencing the agent's behavior.
- Capability inventory: The skill utilizes
Bash ReadandBash Writetools to create the prototype files and executes a Node.js script to process document content. - Sanitization: There is no evidence of content sanitization or validation performed on the text extracted from the input documents before it is used for analysis and code generation.
Audit Metadata