feat: expand advanced elicitation methods with 19 new techniques#2062
feat: expand advanced elicitation methods with 19 new techniques#2062devilliersdutoit wants to merge 9 commits intobmad-code-org:mainfrom
Conversation
Adds 19 new elicitation methods across 7 categories including a new 'framing' category. All existing 50 methods preserved. Entries sorted alphabetically by category then method name. New methods added: - advanced: Chain-of-Thought Scaffolding, Few-Shot Exemplar Priming - collaboration: Six Thinking Hats, Delphi Method - core: Second-Order Thinking, Inversion Analysis, Problem Decomposition, Analogy Mapping, Steelmanning - creative: Constraint Injection, Morphological Analysis - framing (new): Abstraction Laddering, Reframe the Question, Stakeholder Lens Rotation - learning: Deliberate Practice Loop - research: Source Triangulation - risk: Assumption Audit, Cascading Failure Simulation - technical: Boundary & Edge Case Sweep Closes bmad-code-org#2061
are you an AI? |
|
nope, all human
|
|
@devilliersdutoit thanks, this looks interesting, I'm going to think it through a bit on the weekend. Mostly I need to understand how these methods are substantially different from the 50 that are already there. At first blush, it looks like they probably are. Would you be interested in doing a deeper dive into AE? In a nutshell, there is a notion that various situations actually call for specific subsets of AE methods, and it would improve the effectiveness of the whole thing if we explained to LLM what to choose when. And then there is the whole "can we run 2-5 AE methods, with tailored context, same for all methods, or maybe even tailored to the method, and then synthesize the results?" dimension. Human ergonomics could probably be improved as well. |
|
@alexeyv , I would be very intersted in a deep dive into AE. Let me know your thoughts after the weekend. PS: I have to confess that this would the the first time I'm collaborating on any project like this. |
4 participants
What
Expands
src/core-skills/bmad-advanced-elicitation/methods.csvfrom 50 to 69 methods by adding 19 new elicitation techniques and a newframingcategory.Why
The current set is missing several well-established reasoning techniques (Chain-of-Thought Scaffolding, Few-Shot Exemplar Priming, Inversion Analysis, Second-Order Thinking, etc.) that strengthen human-AI collaboration during elicitation sessions.
Fixes #2061
How
methods.csvpreserving all 50 existing entries, sorted alphabetically by categoryframingas a new category (Abstraction Laddering, Reframe the Question, Stakeholder Lens Rotation)advanced,collaboration,core,creative,learning,research,risk, andtechnicalcategoriesTesting
Ran
npm cithen verifiedlint:mdpasses with 0 errors. Confirmed thatformat:check,lint,validate:refs, andvalidate:skillsfailures are identical onmainbefore this change (pre-existing, unrelated to this CSV).