prototype
Pass
Audited by Gen Agent Trust Hub on Jul 9, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it instructs the agent to process untrusted user input to generate and run prototype code. 1. Ingestion points: User prompts used to define prototype logic and names in SKILL.md, LOGIC.md, and UI.md. 2. Boundary markers: No delimiters or ignore instructions are provided to the agent for user data. 3. Capability inventory: The agent is authorized to write files and execute commands via task runners like pnpm, python, and bun (SKILL.md). 4. Sanitization: The instructions do not specify any sanitization or validation of user-provided input.
- [COMMAND_EXECUTION]: The skill directs the agent to execute shell commands using project task runners to run the generated prototypes (SKILL.md), which may lead to command injection if user-controlled strings are not properly escaped.
Audit Metadata