prototype

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to add scripts to the host project's task runners (such as package.json, Makefile, justfile, or pyproject.toml). These scripts are intended to provide a "one command to run" experience for the generated prototypes, which is standard behavior for development-oriented tools.
  • [SAFE]: The instructions explicitly emphasize that prototypes should be "throwaway from day one" and avoid persistence by default, keeping state in memory to minimize risks associated with data handling.
  • [SAFE]: The skill utilizes the disable-model-invocation: true flag in its configuration, which acts as a safety restriction on the agent's behavior during the execution of this specific skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 12:57 PM
Security Audit — agent-trust-hub — prototype