bmad-advanced-elicitation

To refine LLM output, you apply iterative elicitation methods that prompt the model to reconsider and enhance its recently generated content, ensuring higher quality and more accurate results.

Risk assessment on AI-generated text is performed by applying targeted elicitation methods that probe the content for weaknesses and stakeholder misalignment, suggesting specific improvements before finalizing the document.

Yes, you can use iterative elicitation to align LLM content with stakeholder perspectives by selecting specific methods from the registry that probe the generated text for alignment issues and suggest necessary refinements.

A method registry works for collaborative content refinement by offering a diverse set of techniques, integrating with an agent party to iteratively probe, analyze, and improve LLM-generated content for stakeholder alignment.

You need to apply elicitation methods when a complex LLM-generated draft requires deep content analysis, risk assessment, and iterative refinement to identify weaknesses and improve overall output quality.

Limitations of automated elicitation for content improvement include its reliance on recently generated text, meaning it cannot build new content from scratch and requires existing LLM output to function effectively.