using-superpowers

To ensure the correct AI skill is invoked before generating a response, the system uses a conversational guardrail that locates and applies relevant skills via the Skill tool prior to any reply. This grounds interactions in guided capabilities rather than ad hoc reasoning.

Conversational guardrails for AI interaction flow are mechanisms that enforce skill invocation before any response. They guide the system to locate and apply relevant skills at the start of interactions, ensuring replies are grounded in defined capabilities.

You can locate and apply relevant skills at the start of an AI conversation by invoking the appropriate Skill using the Skill tool before replying. If a Skill applies, its output informs the reply; otherwise, standard guidance continues.

Yes, you need to invoke a skill if the user asks clarifying questions. The system gates actions behind appropriate skill usage, ensuring that even clarifying queries are handled by first locating and applying the relevant skill via the Skill tool.

If no applicable skill is found for a user request, the system continues with standard guidance. The process checks for an applicable skill first, and if none exists, it bypasses the skill invocation step to answer directly.