Breakthrough Method for Agile Ai Driven Development
Go to file
Jonah Schulte 2f04fbcd39 feat: add ghost feature detector (reverse gap analysis)
WHO YOU GONNA CALL? 👻 GHOST-FEATURE-BUSTERS!

Implements reverse gap analysis to find "orphaned code" - functionality
that exists in the codebase but has no corresponding story documentation.

Use Case:
"Find functionality that doesn't exist in any story (was vibe coded or
magically appeared) and propose backfilling stories" - User

FEATURE: /ghost-features (trigger: GFD)
---------------------------------------
Scans codebase and cross-references with ALL stories to find orphans:

1. Codebase Scan:
   - React/Vue/Angular components
   - API endpoints (Next.js, NestJS, Express)
   - Database tables (Prisma, TypeORM, migrations)
   - Services and business logic modules
   - Ignores tests, build artifacts, node_modules

2. Cross-Reference with Stories:
   - Check if component mentioned in any File List
   - Check if API mentioned in any Task/AC
   - Check if table mentioned in any story
   - Mark as DOCUMENTED if found, ORPHAN if not

3. Severity Classification:
   - CRITICAL: APIs, auth, payment (security-critical)
   - HIGH: Components, DB tables, services (significant features)
   - MEDIUM: Utilities, helpers
   - LOW: Config files, constants

4. Backfill Story Generation:
   - Analyze orphan code to understand functionality
   - Generate story draft documenting existing implementation
   - Mark most tasks as [x] (already exists)
   - Add tasks for missing tests/docs
   - Suggest epic assignment based on functionality

5. Epic Organization:
   - Option A: Create "Epic-Backfill" for all orphans
   - Option B: Distribute to existing epics
   - Option C: Leave in backlog

Output:
- Orphaned features list (by severity and type)
- Documentation coverage % (what % of code has stories)
- Backfill stories created (if requested)
- Comprehensive report (ghost-features-report-{timestamp}.md)

Usage:
```bash
# Detect orphans in entire sprint
/ghost-features

# Detect orphans in Epic 2
/ghost-features epic_number=2 scan_scope=epic

# Detect and create backfill stories
/ghost-features create_backfill_stories=true

# Full codebase scan
/ghost-features scan_scope=codebase
```

Files Created:
- detect-ghost-features/workflow.yaml: Reverse gap analysis config
- detect-ghost-features/instructions.md: 8-step detection + backfill process

Files Modified:
- sm.agent.yaml: Added GFD (Ghost Feature Detector) menu item

Benefits:
- Prevents "ghost features" from accumulating
- Documents vibe-coded functionality
- Maintains story-code parity
- Enables accurate sprint planning (know what actually exists)
- Makes codebase auditable (every feature has a story)
- Catches manual code additions that bypassed story process
2026-01-07 20:37:19 -05:00
.claude-commands Add batch-super-dev workflow with auto-installation and quality gates 2026-01-07 12:12:13 -05:00
.github feat(docs): Diataxis restructure + Astro/Starlight migration (#1263) 2026-01-07 14:42:15 +08:00
.husky chore: added CC PR review (#871) 2025-11-05 14:14:31 -06:00
.vscode doc cleanup round 1 2025-12-27 18:29:35 +08:00
docs feat: replace batch-and-wait with semaphore pattern for parallel execution 2026-01-07 20:04:39 -05:00
resources/skills feat: add BMAD Guide skill for automatic Claude Code installation 2025-12-27 22:06:37 -05:00
samples/sample-custom-modules agents all indicate hasSidecar true or false, validation requires it, agent builder and validator and editor use the field. Added a better brownfield doc 2025-12-31 21:30:19 +08:00
scripts feat: enhance sprint scripts with smart path resolution and conditional batching 2026-01-06 08:21:44 -05:00
src feat: add ghost feature detector (reverse gap analysis) 2026-01-07 20:37:19 -05:00
test agents all indicate hasSidecar true or false, validation requires it, agent builder and validator and editor use the field. Added a better brownfield doc 2025-12-31 21:30:19 +08:00
tools Merge upstream/main: Diataxis docs restructure + Astro migration 2026-01-07 14:50:08 -05:00
website feat(docs): Diataxis restructure + Astro/Starlight migration (#1263) 2026-01-07 14:42:15 +08:00
.coderabbit.yaml chore: disable CodeRabbit walkthrough (#1115) 2025-12-13 13:34:41 +08:00
.gitignore feat(batch-super-dev): add git commit queue + stricter story validation 2026-01-07 20:10:49 -05:00
.markdownlint-cli2.yaml folder rename from .bmad to _bmad 2025-12-13 16:22:34 +08:00
.npmignore feat: prepare package for npm distribution as @jschulte/bmad-method 2025-12-27 22:57:45 -05:00
.npmrc feat: v6.0.0-alpha.0 - the future is now 2025-09-28 23:17:07 -05:00
.nvmrc feat: v6.0.0-alpha.0 - the future is now 2025-09-28 23:17:07 -05:00
.prettierignore folder rename from .bmad to _bmad 2025-12-13 16:22:34 +08:00
ASSESSMENT_SUMMARY.txt fix: resolve critical complexity routing issues from multi-agent review 2026-01-07 16:34:44 -05:00
CHANGELOG.md chore: bump version to 6.1.0-alpha.3 2026-01-07 20:13:39 -05:00
CNAME Add CNAME file 2026-01-07 18:18:12 +08:00
CONTRIBUTING.md docs: add how-to guide for getting BMAD answers + fix Discord channels (#1265) 2026-01-07 16:05:05 +08:00
FEATURE-SUMMARY.md refactor: use invoke-workflow for orchestration 2025-12-26 20:26:30 -05:00
INTEGRATION-NOTES.md docs: update all documentation for super-dev-pipeline 2025-12-28 12:23:55 -05:00
LICENSE quick update output modified 2025-12-15 17:30:12 +08:00
PR-DESCRIPTION.md feat: add push-all workflow as standalone command 2025-12-26 18:32:00 -05:00
PR-STORY-PIPELINE.md docs: update all documentation for super-dev-pipeline 2025-12-28 12:23:55 -05:00
README-changes.md docs: update README-changes.md with v1.3.0 complexity routing feature 2026-01-07 16:38:59 -05:00
README.md fix install for alpha instruction in the readme 2025-12-29 15:45:38 +08:00
TESTING-GUIDE.md refactor: use invoke-workflow for orchestration 2025-12-26 20:26:30 -05:00
eslint.config.mjs feat(docs): Diataxis restructure + Astro/Starlight migration (#1263) 2026-01-07 14:42:15 +08:00
package-lock.json chore: bump version to 6.1.0-alpha.3 2026-01-07 20:13:39 -05:00
package.json chore: bump version to 6.1.0-alpha.3 2026-01-07 20:13:39 -05:00
prettier.config.mjs feat: v6.0.0-alpha.0 - the future is now 2025-09-28 23:17:07 -05:00

README.md

BMad Method & BMad Core

Stable Version Alpha Version License: MIT Node.js Version Discord


🎉 NEW: BMAD V6 Installer - Create & Share Custom Content!

The completely revamped BMAD V6 installer now includes built-in support for creating, installing, and sharing custom modules, agents, workflows, templates, and tools! Build your own AI solutions or share them with your team - and real soon, with the whole BMad Community througha verified community sharing portal!

What's New:

  • 📦 Streamlined Custom Module Installation - Package your custom content as installable modules
  • 🤖 Agent & Workflow Sharing - Distribute standalone agents and workflows
  • 🔄 Unitary Module Support - Install individual components without full modules
  • ⚙️ Dependency Management - Automatic handling of module dependencies
  • 🛡️ Update-Safe Customization - Your custom content persists through updates

📚 Learn More:


AI-Driven Agile Development That Scales From Bug Fixes to Enterprise

Build More, Architect Dreams (BMAD) with 21 specialized AI agents across 4 official modules, and 50+ guided workflows that adapt to your project's complexity—from quick bug fixes to enterprise platforms, and new step file workflows that allow for incredibly long workflows to stay on the rails longer than ever before!

Additionally - when we say 'Build More, Architect Dreams' - we mean it! The BMad Builder has landed, and now as of Alpha.15 is fully supported in the installation flow via NPX - custom stand along agents, workflows and the modules of your dreams! The community forge will soon open, endless possibility awaits!

🚀 v6 is a MASSIVE upgrade from v4! Complete architectural overhaul, scale-adaptive intelligence, visual workflows, and the powerful BMad Core framework. v4 users: this changes everything. See what's new →

📌 v6 Alpha Status: Near-beta quality with vastly improved stability. Documentation is being finalized. New videos coming soon to BMadCode YouTube.

🎯 Why BMad Method?

Unlike generic AI coding assistants, BMad Method provides structured, battle-tested workflows powered by specialized agents who understand agile development. Each agent has deep domain expertise—from product management to architecture to testing—working together seamlessly.

Key Benefits:

  • Scale-Adaptive Intelligence - Automatically adjusts planning depth from bug fixes to enterprise systems
  • Complete Development Lifecycle - Analysis → Planning → Architecture → Implementation
  • Specialized Expertise - 19 agents with specific roles (PM, Architect, Developer, UX Designer, etc.)
  • Proven Methodologies - Built on agile best practices with AI amplification
  • IDE Integration - Works with Claude Code, Cursor, Windsurf, VS Code

🏗️ The Power of BMad Core

BMad Method is actually a sophisticated module built on top of BMad Core (Collaboration Optimized Reflection Engine). This revolutionary architecture means:

  • BMad Core provides the universal framework for human-AI collaboration
  • BMad Method leverages Core to deliver agile development workflows
  • BMad Builder lets YOU create custom modules as powerful as BMad Method itself

With BMad Builder, you can architect both simple agents and vastly complex domain-specific modules (legal, medical, finance, education, creative) that will soon be sharable in an official community marketplace. Imagine building and sharing your own specialized AI team!

📊 See It In Action

BMad Method Workflow

Complete BMad Method workflow showing all phases, agents, and decision points

🚀 Get Started in 3 Steps

1. Install BMad Method

# Install v6 RECOMMENDED
npx bmad-method@alpha install
# Install v4 Legacy (not recommended if starting fresh)
npx bmad-method install
# OR
npx bmad-method@latest install

2. Initialize Your Project

Load any agent in your IDE and run:

*workflow-init

This analyzes your project and recommends the right workflow track.

3. Choose Your Track

BMad Method adapts to your needs with three intelligent tracks:

Track Use For Planning Time to Start
Quick Flow Bug fixes, small features Tech spec only < 5 minutes
📋 BMad Method Products, platforms PRD + Architecture + UX < 15 minutes
🏢 Enterprise Compliance, scale Full governance suite < 30 minutes

Not sure? Run *workflow-init and let BMad analyze your project goal.

🔄 How It Works: 4-Phase Methodology

BMad Method guides you through a proven development lifecycle:

  1. 📊 Analysis (Optional) - Brainstorm, research, and explore solutions
  2. 📝 Planning - Create PRDs, tech specs, or game design documents
  3. 🏗️ Solutioning - Design architecture, UX, and technical approach
  4. Implementation - Story-driven development with continuous validation

Each phase has specialized workflows and agents working together to deliver exceptional results.

🤖 Meet Your Team

12 Specialized Agents working in concert:

Development Architecture Product Leadership
Developer Architect PM Scrum Master
UX Designer Test Architect Analyst BMad Master
Tech Writer

Test Architect integrates with @seontechnologies/playwright-utils for production-ready web app fixture-based utilities.

Each agent brings deep expertise and can be customized to match your team's style.

📦 What's Included

Official Modules

  • BMad Method (BMM) - Complete agile development framework

    • 12 specialized agents
    • 34 workflows across 4 phases
    • Stand Along Quick Spec Flow for a streamlined simple implementation process
    • → Documentation Hub
  • BMad Builder (BMB) - Create custom agents and workflows

    • Build anything from simple agents to complex modules
    • Create domain-specific solutions (legal, medical, finance, education)
    • → Builder Guide
  • Creative Intelligence Suite (CIS) - Innovation & problem-solving

Key Features

  • 🎨 Customizable Agents - Modify personalities, expertise, and communication styles
  • 🌐 Multi-Language Support - Separate settings for communication and code output
  • 📄 Document Sharding - 90% token savings for large projects
  • 🔄 Update-Safe - Your customizations persist through updates
  • 🚀 Web Bundles - Use in ChatGPT, Claude Projects, or Gemini Gems

📚 Documentation

For v4 Users

💬 Community & Support

🛠️ Development

If you would like to contribute, first check the CONTRIBUTING.md for full development guidelines.

What's New in v6

v6 represents a complete architectural revolution from v4:

🚀 Major Upgrades

  • BMad Core Framework - Modular architecture enabling custom domain solutions
  • Scale-Adaptive Intelligence - Automatic adjustment from bug fixes to enterprise
  • Visual Workflows - Beautiful SVG diagrams showing complete methodology
  • BMad Builder Module - Create and share your own AI agent teams
  • 50+ Workflows - Up from 20 in v4, covering every development scenario
  • 19 Specialized Agents - Enhanced with customizable personalities and expertise
  • Update-Safe Customization - Your configs persist through all updates
  • Web Bundles - Use agents in ChatGPT, Claude, and Gemini
  • Multi-Language Support - Separate settings for communication and code
  • Document Sharding - 90% token savings for large projects

🔄 For v4 Users

📄 License

MIT License - See LICENSE for details.

Trademarks: BMad™ and BMAD-METHOD™ are trademarks of BMad Code, LLC.

Supported by:  DigitalOcean


Contributors

Built with ❤️ for the human-AI collaboration community