bmad-advanced-elicitation

You can refine LLM outputs by applying iterative elicitation methods such as socratic, first principles, pre-mortem, and red team techniques. This approach pushes the LLM to reconsider and enhance its generated content through in-depth evaluation and critique.

For deeper critique of LLM-generated content, you can use socratic, first principles, pre-mortem, and red team elicitation methods. These techniques provide thorough analysis by forcing the LLM to iteratively reconsider and improve its outputs.

Use pre-mortem and red team analysis when you need a deeper critique or more refined output from an LLM. These iterative elicitation methods support in-depth evaluation, pushing the model to identify potential flaws and enhance its analysis.

Yes, you can apply first principles thinking to refine LLM analysis without external coding dependencies. The skill operates through scripts to implement iterative elicitation methods, requiring no additional setup to push the LLM to reconsider its outputs.

The best way to push an LLM to reconsider its initial output is through iterative elicitation methods like socratic questioning and red team analysis. These techniques systematically challenge the model's reasoning to achieve a more refined and enhanced response.