Skip to content
Kitbay
Skill

bmad-advanced-elicitation

MITUpdated on Nov 11, 202655,252GitHub starsIncludes scripts

A skill from the BMad Method that pushes a draft beyond its first answer by running several independent critique passes and proposing one revision.

What the skill teaches

A skill from the BMad Method that pushes a draft beyond its first answer by running several independent critique passes and proposing one revision. The target is whatever the user or a calling skill points at, or else the last answer in the conversation. The agent picks named methods from a catalogue, or by default one method aimed at the likely weak spot, one unpredictable method and an open critique, and runs each in a fresh subagent with a short brief so the readers stay unbiased. Their located findings are merged into a single revision, set-aside findings are listed with reasons, and the user applies all, some or none before another round. Each round is logged to a report folder.

Capabilities

  • Choose critique methods for a draft
  • Run independent critiques in subagents
  • Merge findings into one revision
  • Let the user accept changes by number