prototype
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to launch prototypes using existing project task runners such as
pnpm,python, orbun, relying on the project's own command configurations.\n- [REMOTE_CODE_EXECUTION]: The skill involves the generation and subsequent execution of standalone HTML/JS files and UI components. To manage risk, it instructs that logic be kept in pure, liftable modules and that UI prototypes be gated by environment checks (e.g.,process.env.NODE_ENV !== 'production') to prevent accidental shipment to production.\n- [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it translates user inputs and project files into executable code.\n - Ingestion points: User-provided design questions and existing project source code as specified in
SKILL.md.\n - Boundary markers: Absent; the instructions do not require the use of delimiters or 'ignore' warnings to protect against malicious content in the inputs.\n
- Capability inventory: The agent has the ability to write files and execute shell commands via project-standard task runners.\n
- Sanitization: Absent; no specific input escaping or validation is mandated before the data is used to generate the prototype code.
Audit Metadata