html-prototype
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the execution of shell commands to manage dependencies via
npm installand to run theextract_pdf_text.jsscript for processing PDF files. - [EXTERNAL_DOWNLOADS]: The skill downloads the
pdf-parselibrary from the NPM registry during its setup phase. - [PROMPT_INJECTION]: An indirect prompt injection surface is present because the skill parses untrusted user-provided documents (PDF, Markdown, or Text) to generate executable code without sanitization.
- Ingestion points: User-specified files and the
screen-design/directory. - Boundary markers: No delimiters or instructions are used to distinguish untrusted data from the agent's generation logic.
- Capability inventory: The skill uses Bash tools to read local files and write HTML/JS prototypes.
- Sanitization: The skill does not perform validation or sanitization on text extracted from source documents before code generation.
Audit Metadata