- Fix install-bmad.sh line ending issues preventing execution on Git Bash/WSL
- Add comprehensive error handling and Windows path normalization
- Create install-bmad.bat for native Windows Command Prompt support
- Update .gitignore to exclude generated .bmad/ and exam/ directories
- Add detailed installation troubleshooting guide to README
- Improve installation verification with better error messages and fallback options
Resolves installation failures across different platforms and environments.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Updated installation approach to use local repository paths instead of incorrect GitHub links:
**New Local Installation Script (install-bmad.sh):**
- Copies complete BMAD system from local repository to project
- Installs both bmad-system/ and bmad-agent/ directories
- Creates CLAUDE.md reference file in project root
- Generates project-specific configuration
- Automatic installation verification
- Clear next steps guidance
**Installation Process:**
1. Copy install-bmad.sh from BMAD repository to project directory
2. Run ./install-bmad.sh to install complete BMAD system
3. Creates .bmad/ directory with all system files and agents
4. Creates CLAUDE.md reference file for persistent Claude Code integration
**Files Copied to Projects:**
- Complete bmad-system/ documentation and modules
- Complete bmad-agent/ personas, tasks, checklists, templates
- Project-specific configuration (project-config.yml)
- CLAUDE.md reference guide for Claude Code sessions
**Updated Documentation:**
- README.md: Local installation instructions instead of GitHub links
- bmad-system/README.md: Complete installation process with file structure
- CLAUDE.md: Persistent reference guide for Claude Code sessions
**Usage Pattern:**
1. Download/clone BMAD-METHOD repository locally
2. Copy install-bmad.sh to your project
3. Run installation script
4. Reference CLAUDE.md in Claude Code sessions
5. Claude reads .bmad/ files for BMAD-powered development
The Enhanced BMAD System now has a complete local installation
system that works correctly with downloaded repositories.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
This commit completes the comprehensive 4-phase Enhanced BMAD System transformation:
**Phase 4 Implementation:**
- Self-Optimization Engine: Meta-optimization and adaptive algorithms
- Enterprise Architecture Platform: Enterprise-scale architectural governance
- Advanced Governance Framework: Comprehensive governance and compliance
- Strategic Intelligence Dashboard: Executive-level insights and analytics
- Enterprise Security & Compliance: Zero-trust security and automated compliance
- Advanced Monitoring & Analytics: AI-powered monitoring and observability
- Cost Optimization Engine: Financial intelligence and cost optimization
**Complete System Features:**
- 27 comprehensive system modules across all 4 phases
- Autonomous development with 4 levels of autonomy
- Universal LLM compatibility (Claude, GPT-4, Gemini, DeepSeek, Llama)
- Enterprise-scale governance, security, and compliance automation
- AI-powered analytics, optimization, and self-improvement
- Comprehensive monitoring, alerting, and automated remediation
The Enhanced BMAD System is now a complete enterprise-grade, self-optimizing,
intelligent development platform that transforms Claude Code capabilities
throughout the entire software development lifecycle.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* docs: add headers and improve formatting for BMAD orchestrator agent documentation
## CHANGES
- Add configuration header to cfg file
- Improve numbered list formatting consistency
- Add proper heading punctuation throughout
- Enhance readability with cleaner structure
- Standardize markdown formatting conventions
* gitignore update
* Plaform Engineer role for a robust infrastructure (#135)
* Add Platform Engineer role to support a robust and validated infrastructure
* Platform Engineer and Architect boundaries, confidence levels, domain expertise
* remove duplicate task, leftover artifact
* Consistency, workflow, feedback loops between architect and PE
* PE customization generalized, updated Architect, consistency check
* style: add VSCode integration and standardize document formatting
CHANGES
- Introduce VSCode recommended extensions and project-specific settings.
- Update `.gitignore` to track the `.vscode` directory.
- Apply consistent markdown formatting to all checklist documents.
- Standardize spacing, list styles, and headers in personas.
- Refine formatting and sectioning in task definition files.
- Ensure newline termination for all modified text files.
- Correct code block specifiers and minor textual content.
* docs: remove exclamation from header
* fix: spacing at end of line
---------
Co-authored-by: Brian Madison <brianmadison@Brians-MacBook-Pro.local>
Co-authored-by: Sebastian Ickler <icklers@users.noreply.github.com>
* docs: update README links to renamed orchestrator config files
## CHANGES
- Update orchestrator link to ide-bmad-orchestrator.md
- Update config link to ide-bmad-orchestrator.cfg.md
- Fix Markdownlint errors in ide-bmad-orchestrator.cfg.md
* docs: update path references from BETA-V3 to bmad-agent in instruction.md
## CHANGES
- Update all BETA-V3 directory references to bmad-agent
- Clarify paths are relative to bmad-agent directory
- Fix markdown formatting for numbered lists
- Update IDE orchestrator configuration path examples
- Correct agent-root path in Data Resolution section
- Remove references to deprecated BETA-V3 structure
Quick note about previous repo that this one replaces:
This method is a full replacement and enhancement to what was hinted at and partially described in the custom-agents-rules-generator [this repo](https://github.com/bmadcode/cursor-custom-agents-rules-generator). This is now more tailored to being generic and working with any IDE (not just cursor specific) and the custom rule used to generate rules is no longer needed in Cursor anyways as of 0.49x (And the other IDE's now support auto rule generation also) and with custom agents and agile artifacts, rules become less necessary. Rules that apply to general standards can be build into your developer agents. For example, you can expand the dev persona agent herein to be a typescript dev agent, or a python dev agent, or even a ui dev agent - all with the best practices you want it to follow baked in! By having multiple dev types, you can have specialized devs with the rules in their context primed for what they will be working on - instead of overall bloated rules that do not apply to every task at hand. This is all optional, but you can start to see why this replaces the detailed rules based workflows.
Where IDE rules will still apply, is for fine tuning quick one off rules as you are going if you find the agent making many mistakes in certain ways. In the future you can craft this adherance into your agile artifacts and stories, or the custom mode configurations!