The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requires access to the Bash and Task tools to scaffold projects and configure development environments. This allows for full shell access, which is consistent with the skill's purpose but represents a significant capability.\n- [EXTERNAL_DOWNLOADS]: The agent integrates various well-known APIs and libraries from trusted services like OpenAI, Stripe, and Supabase. These external resources are standard for modern application prototyping.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted user instructions to generate and execute code.\n
Ingestion points: User requirements for app prototypes and feature descriptions.\n
Boundary markers: No delimiters or specific instructions are provided to the agent to ignore potentially malicious embedded instructions in user data.\n
Capability inventory: The agent has high-privilege tool access, including Bash, Write, and Task, which can be leveraged if an injection is successful.\n
Sanitization: There is no indication that user-provided input is sanitized before being used to generate code or execute system commands.

rapid-prototyper