Get the FREE Ultimate OpenClaw Setup Guide →

bmad-creative-intelligence

Scanned
npx machina-cli add skill xmm/codex-bmad-skills/bmad-creative-intelligence --openclaw
Files (1)
SKILL.md
3.2 KB

BMAD Creative Intelligence

Trigger Intents

  • bmad:ideate
  • bmad:research-deep

Workflow Variants

  1. ideate
  • Generate and prioritize option sets for a defined problem.
  1. research-deep
  • Produce evidence-backed analysis with sources and recommendations.

Inputs

  • problem statement and constraints
  • existing BMAD artifacts in docs/bmad/
  • target market or technical domain

Language Guard (Mandatory)

Enforce language selection separately for chat responses and generated artifacts.

Chat language (communication_language) fallback order:

  1. language.communication_language from bmad/project.yaml
  2. English

Rules for chat responses:

  • Use the resolved chat language for all assistant responses (questions, status updates, summaries, and handoff notes).
  • Do not switch chat language unless the user explicitly requests a different language in the current thread.

Artifact language (document_output_language) fallback order:

  1. language.document_output_language from bmad/project.yaml
  2. English

Rules for generated artifacts:

  • Use the resolved artifact language for all generated BMAD documents and structured artifacts.
  • write prose and field values in the resolved document language
  • avoid mixed-language requirement clauses with English modal verbs (for example, System shall followed by non-English text)
  • allow English acronyms/abbreviations in non-English sentences (for example, API, SLA, KPI, OAuth, WCAG)
  • Keep code snippets, CLI commands, file paths, and identifiers in their original technical form.

Mandatory Reference Load

Before executing ideate or research-deep, read REFERENCE.md first. Treat REFERENCE.md as required context for method selection and output quality.

Output Contract

  • ideation output -> docs/bmad/brainstorm.md
  • deep research output -> docs/bmad/research-deep.md
  • optional option memo -> docs/bmad/options.md

Core Workflow

  1. Select method based on uncertainty and decision type.
  2. Generate options or gather evidence.
  3. Evaluate impact, effort, risk, and confidence.
  4. Produce ranked recommendations.
  5. Suggest concrete next BMAD intent.

Script Selection

  • SCAMPER prompts: 
    bash scripts/scamper-prompts.sh
  • SWOT structure: 
    bash scripts/swot-template.sh
  • Research source collection: 
    bash scripts/research-sources.sh

Template Map

  • templates/brainstorm-session.template.md

  • Why: structure ideation sessions and ranking output.

  • templates/research-report.template.md

  • Why: structure evidence collection and recommendations.

Reference Map

  • REFERENCE.md

  • Must read first for method selection and workflow guidance.

  • resources/brainstorming-techniques.md

  • Use for ideation frameworks and facilitation patterns.

  • resources/research-methods.md

  • Use for deep research design and evidence quality.

Quality Gates

  • alternatives are concrete and comparable
  • assumptions and confidence levels are explicit
  • major claims include source context or uncertainty notes
  • final recommendation is actionable

Source

git clone https://github.com/xmm/codex-bmad-skills/blob/main/skills/bmad-creative-intelligence/SKILL.mdView on GitHub

Overview

BMAD Creative Intelligence blends structured ideation with rigorous deep research to generate option sets and evidence-backed insights for BMAD projects. It supports two workflow modes: ideate for generating and ranking options, and research-deep for thorough analysis with sources and concrete recommendations. It relies on clear inputs (problem statements, existing BMAD docs, and domain context) and requires reading REFERENCE.md before use.

How This Skill Works

Select a method based on uncertainty and decision type. The system either generates and prioritizes option sets or collects evidence, then evaluates impact, effort, risk, and confidence. It outputs ranked recommendations and a concrete next BMAD intent, and writes artifacts to the BMAD docs (docs/bmad/brainstorm.md for ideate or docs/bmad/research-deep.md for research-deep). A mandatory Reference Load (REFERENCE.md) guides method choice and quality.

When to Use It

  • When you need to generate and prioritize option sets for a defined BMAD problem.
  • When you require an evidence-backed analysis with sources and recommendations.
  • When uncertainty is high and you want explicit assumptions and confidence notes.
  • When you need actionable next BMAD intents to guide backlog or roadmap.
  • When you need a concrete comparison of alternatives with impact, effort, and risk.

Quick Start

  1. Step 1: Read REFERENCE.md and collect inputs (problem, constraints, docs/bmad artifacts, domain).
  2. Step 2: Run the appropriate workflow (ideate or research-deep) to generate options or evidence.
  3. Step 3: Review outputs and export to docs/bmad/brainstorm.md or docs/bmad/research-deep.md (and optionally docs/bmad/options.md).

Best Practices

  • Frame with a concrete problem statement and constraints.
  • Ensure alternatives are concrete and comparable.
  • Make assumptions explicit and assign confidence levels.
  • Cite sources or context for major claims and mark uncertainties.
  • Export outputs to the proper BMAD docs paths using the provided templates.

Example Use Cases

  • Brainstorm and rank feature options for a BMAD module.
  • Deep-dive research on a new domain with sources and concrete recommendations.
  • Produce an options memo comparing design alternatives with pros/cons and tradeoffs.
  • Generate actionable next BMAD intents to seed the backlog.
  • Deliver an evidence-backed risk and impact analysis for a regulatory or compliance requirement.

Frequently Asked Questions

Add this skill to your agents
Sponsor this space

Reach thousands of developers ↗