760a338af3
Add role-optimized LLM settings to all BMAD agents
...
AGENT-SPECIFIC LLM OPTIMIZATION:
🔧 Development Agents (Precision Focus):
- Dev (James): temp=0.4, top_p=0.85 - Low temp for precise code generation
- QA (Quinn): temp=0.3, top_p=0.8 - Very low temp for systematic analysis
🏗️ Technical Architecture (Balanced Creativity):
- Architect (Winston): temp=0.6, top_p=0.9 - Moderate creativity for design solutions
- SM (Bob): temp=0.5, top_p=0.85 - Structured story creation consistency
💡 Creative & Strategic (Higher Innovation):
- Analyst (Mary): temp=0.8, top_p=0.95 - High creativity for business ideation
- UX Expert (Sally): temp=0.75, top_p=0.92 - Creative design thinking
📋 Product Management (Balanced Structure):
- PM (John): temp=0.7, top_p=0.9 - Strategic creativity with structure
- PO (Sarah): temp=0.6, top_p=0.88 - User-focused requirements balance
🎭 Orchestration (Guidance Focused):
- BMad Orchestrator: temp=0.7, top_p=0.9 - Clear workflow coordination
REASONING INCLUDED:
Each agent includes detailed reasoning for their specific LLM parameter choices
based on their role requirements, output precision needs, and creative demands.
EXPECTED BENEFITS:
- More consistent code generation from development agents
- Enhanced creative output from ideation-focused agents
- Balanced technical creativity for architecture decisions
- Optimized agent performance for their specific use cases
2025-07-22 11:01:13 -04:00
3182aa9f8d
Implement automatic remediation execution in reality audit
...
Major Enhancement: Auto-Execute Remediation Instead of Just Recommending
CHANGES:
- Enhanced Phase 9: Added automatic scope issue detection and remediation execution
- Added SCOPE_REMEDIATION_NEEDED trigger for oversized stories (>8 tasks, >25 subtasks)
- Added mixed concerns detection (implementation + integration in same story)
- Automatic execution of create-remediation-story.md when triggers detected
- Enhanced QA agent with execution workflow and trigger conditions
- Added enhanced_auto_remediation section with complete workflow sequence
WORKFLOW IMPROVEMENT:
- Before: Analyze → Recommend → User runs *create-remediation → Get solution
- After: Analyze → Auto-execute → Present completed remediation stories
TRIGGER CONDITIONS:
- Reality score < 80
- Build/runtime failures
- Story tasks > 8 or subtasks > 25
- Mixed implementation/integration concerns
- Critical simulation patterns > 3
ELIMINATES USER CONFUSION:
- No more 'run this command next' recommendations
- Automatic story splitting when scope issues detected
- Immediate surgical remediation story generation
- Ready-to-use options presented automatically
This transforms the reality audit from an analysis tool into a complete
remediation execution system.
2025-07-21 16:09:42 -04:00
74b6c489f5
Add automatic options presentation to reality audit
...
- Adds Phase 10: Automatic Next Steps Presentation
- Presents 4 grade-based option sets (A/B/C/D-F) with specific recommendations
- Eliminates user confusion about 'what's next' after audit results
- Provides immediate next commands based on reality score and issues
- Includes effort estimates and specific action items for each option
- Enhances user experience by removing need to ask 'what are my options'
- Covers all scenarios from excellent quality to significant issues
2025-07-21 15:53:14 -04:00
86743bb3e6
Optimize reality audit to use Grep tool instead of bash commands
...
- Replace bash grep loops with Grep tool usage for pattern detection
- Eliminates need for multiple command approvals during QA audit
- Maintains all simulation pattern detection capabilities
- Improves user experience by reducing workflow interruptions
- Updated task instructions to guide agents to use Grep tool
- Preserves comprehensive scoring and reporting functionality
2025-07-21 12:02:58 -04:00
1af89a3e69
Add Task Execution Enforcement feature to BMAD Quality Framework
...
- Enhanced enhancements.md with eighth game-changing feature
- Added task-execution-validation.md checklist for reliable automation
- Updated dev and qa agents with task execution enforcement
- Prevents automation workflow failures through mandatory file validation
- Ensures agents use Read tool instead of generic Task tool
- Includes comprehensive troubleshooting and validation steps
2025-07-21 11:52:10 -04:00
4c2b86d051
Add dual-track progress enforcement to prevent story file update failures
...
- Enhanced dev agent with mandatory dual-track updates (story file + TodoWrite)
- Added validation gates to halt progression if story file not updated
- Prevents completed stories showing unchecked tasks
- Updated documentation with dual-track progress feature
Fixes: Story completion without incremental checkbox updates
2025-07-21 11:18:05 -04:00
e57448b652
enhancements
2025-07-21 09:43:38 -04:00
4c6adf52d3
removing backups
2025-07-21 09:19:06 -04:00
cb66340738
Fix missing *develop-story command preventing automatic task completion
...
- Added develop-story to commands list in dev agent
- Enables systematic workflow execution with automatic progress tracking
- Tasks now properly marked [x] as agent progresses through story implementation
2025-07-21 09:05:16 -04:00
3faba78db0
Add automatic quality framework with loop detection and external LLM collaboration
...
The key elements captured:
- Enterprise-grade quality framework (main feature)
- Automatic loop detection (your key innovation)
- External LLM collaboration (the copy-paste prompt solution)
- Zero-touch workflow (eliminates manual handoffs)
2025-07-21 06:08:13 -04:00
df57d772ca
feat: Installation includes a getting started user guide with detailed mermaid diagram
2025-07-19 21:18:06 -05:00
49e489701e
fix: lint fix
2025-07-19 10:07:34 -05:00
c6dc345b95
direct commands in agents
2025-07-19 00:30:42 -05:00
3367fa18f7
version alignment
2025-07-19 00:04:16 -05:00
d0a8c581af
fixed roomodes double bmad
2025-07-16 23:36:24 -05:00
0723eed881
fix: bundle build
2025-07-14 00:07:29 -05:00
3621c330e6
fix: improve agent task folowing in agressing cost saving ide model combos
2025-07-14 00:06:25 -05:00
afea271e5e
fix: annoying YAML lint error
2025-07-13 15:51:46 -05:00
6b860bfee4
improve agent performance in claude code slash commands
2025-07-13 11:53:22 -05:00
f62c05ab0f
fix: brianstorming facilitation output
2025-07-13 11:33:06 -05:00
e9e541a52e
feat: Claude Code slash commands for Tasks and Agents!
2025-07-12 21:08:13 -05:00
24a35ff2c4
core agents alignment
2025-07-12 20:16:05 -05:00
3c13c56498
feat: bmad-master can load kb properly
2025-07-12 10:27:21 -05:00
db302309f4
fix: installer improvement
2025-07-07 22:11:32 -05:00
cadf8b6750
fix: installer for github copilot asks follow up questions right away now so it does not seem to hang, and some minor doc improvements
2025-07-07 20:46:55 -05:00
1b86cd4db3
fix: doc updates
2025-07-06 20:51:40 -05:00
a5ffe7b9b2
fix: improve output
2025-07-06 20:40:08 -05:00
2601fa7205
version bump
2025-07-06 19:54:46 -05:00
92201ae7ed
fix: build web bundles with new file extension includsion
2025-07-06 19:39:34 -05:00
746ba573fa
specify md ot yaml
2025-07-06 18:26:09 -05:00
339745c3f3
combine startup with activation in agent files
2025-07-06 16:07:39 -05:00
b78537115d
feat: experimental doc creator v2 and template system
2025-07-06 12:11:55 -05:00
0a61d3de4a
fix: readme consolidation and version bumps
2025-07-06 11:13:09 -05:00
4e03f8f982
merge conflicts resolved
2025-07-06 10:34:53 -05:00
5fc69d773a
web build optimization
2025-07-06 10:32:39 -05:00
9e6940e8ee
refactor: Standardize on 'GitHub Copilot' branding ( #296 )
...
* refactor: Standardize on 'GitHub Copilot' branding
- Update all references from 'Github Copilot' to 'GitHub Copilot' (official branding)
- Simplify GitHub Copilot guide reference in README
- Rebuild distribution files to reflect changes
- Ensure consistent branding across documentation and configuration
* fix: add Trae IDE support while maintaining GitHub Copilot branding
2025-07-06 08:49:22 -05:00
c107af0598
feat: WIP create-docv2
2025-07-06 00:10:00 -05:00
be9453f234
Merge branch 'main' of github.com:bmadcode/BMAD-METHOD
2025-07-05 23:19:45 -05:00
47b9d9f3e8
fix: spelling errors in documentation. ( #297 )
...
* fix: correct typos in documentation and agent files
Fix multiple instances of "assest" typo to "assets" in documentation
Correct "quetsions" typo to "questions" in repository structure sections
Add new words to cSpell dictionary in VS Code settings
* feat(trae): add support for trae ide integration
- Add trae guide documentation
- Update installer to support trae configuration
- Include trae in ide options and documentation references
- Fix typo in architect agent documentation
---------
Co-authored-by: Devin Stagner <devin@blackstag.family>
2025-07-05 21:08:26 -05:00
1bc9960808
feat: big improvement to advanced elicitation
2025-07-05 14:01:29 -05:00
9f53caf4c6
feat: Massive improvement to the brainstorming task!
2025-07-04 23:36:18 -05:00
42684e68af
fix: update web bundles
2025-07-04 21:31:52 -05:00
3520fae655
minor updates
2025-07-04 21:27:50 -05:00
2874a54a9b
feat: improvements to agent task usage, sm story drafting, dev implementation, qa review process, and addition of a new sm independant review of a draft story
2025-07-04 21:18:16 -05:00
1c845e5b2c
fix: version bump and web build fix
2025-07-04 12:39:17 -05:00
094f9f3eab
fix: yaml standardization in files and installer actions
2025-07-04 11:53:57 -05:00
38dd71db7f
doc reference changes from vs-code-copilot to github-copilot
2025-07-04 08:04:27 -05:00
f440d14565
doc and text cleanup
2025-07-04 07:47:57 -05:00
87c451a5c3
fix: version bump and restore dist folder
2025-07-02 23:14:54 -05:00
a4c0b1839d
fix: standardize on file extension .yaml instead of a mix of yml and yaml
2025-07-02 19:59:49 -05:00
James (Claude Code)
DevForgeAI
Brian Madison
David Elisma
manjaroblack