bmad-advanced-elicitation

Refine LLM output through deterministic validation and refinement steps.

Updated Mar 18, 2026
One-click install
npx skills add https://github.com/robotics-playground/skills --skill bmad-advanced-elicitation-robotics-playground
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-advanced-elicitation
Source: https://github.com/robotics-playground/skills/tree/main/_bmad/core/skills/bmad-advanced-elicitation
Command: npx skills add https://github.com/robotics-playground/skills --skill bmad-advanced-elicitation-robotics-playground

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Push the LLM to reconsider, refine, and improve its recent output.

Core Features & Use Cases

  • Deterministic multi-step prompting: enforces a fixed sequence of validation and refinement steps.
  • Iterative improvement: supports multiple cycles of evaluation and revision to enhance quality.
  • Safe elicitations: includes safety checks to avoid prompt injection or leakage of system prompts.

Quick Start

Provide the initial output you want refined, and initiate the elicitation workflow.

Frequently Asked Questions about bmad-advanced-elicitation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I refine LLM outputs through iterative prompt engineering?

You refine LLM outputs through iterative prompt engineering by applying a deterministic elicitation workflow that pushes the model to reconsider and improve its recent response across multiple validation checkpoints.

What is iterative elicitation for long-form responses?

Iterative elicitation for long-form responses is a multi-step refinement process that enforces mandatory validation cycles, ensuring the LLM evaluates and revises its output to enhance overall quality and accuracy.

How do I validate complex instructions and prevent prompt injection during elicitation?

You validate complex instructions and prevent prompt injection during elicitation by using a workflow that enforces safety checks at validation checkpoints, avoiding system prompt leakage while refining outputs.

What's the best way to force an LLM to reconsider its recent output?

The best way to force an LLM to reconsider its recent output is to use a structured elicitation workflow with a fixed sequence of refinement steps, supporting multiple cycles of evaluation and revision.

Does iterative elicitation work for error-prone LLM outputs?

Yes, iterative elicitation works for error-prone LLM outputs by applying deterministic multi-step prompting that enforces mandatory validation checkpoints to catch and correct errors during the refinement cycles.