prototype
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions involve interpreting user prompts to generate functional code artifacts (HTML files or UI components). \n
- Ingestion points: User prompts are used to determine the question being answered and the shape of the prototype (SKILL.md).\n
- Boundary markers: There are no instructions to use delimiters or ignore potentially malicious embedded instructions in the user's prompt or surrounding code.\n
- Capability inventory: The skill directs the agent to write files to the local system (HTML for logic demos, TSX/JSX for UI variants).\n
- Sanitization: No sanitization or validation of the input prompt is described before the agent generates and writes the prototype code.\n- [COMMAND_EXECUTION]: The skill recommends using local task runners like pnpm, python, or bun to execute prototypes. These are standard developer tools and the instructions are limited to running local project tasks.
Audit Metadata