chore(release): 4.23.0 [skip ci]

# [4.23.0](https://github.com/bmadcode/BMAD-METHOD/compare/v4.22.1...v4.23.0) (2025-07-01)

### Features

* VS Code Copilot integration ([#284](https://github.com/bmadcode/BMAD-METHOD/issues/284)) ([1a4ca4f](1a4ca4ffa6))

CHANGELOG.md

@@ -1,3 +1,17 @@
+# [4.23.0](https://github.com/bmadcode/BMAD-METHOD/compare/v4.22.1...v4.23.0) (2025-07-01)
+
+
+### Features
+
+* VS Code Copilot integration ([#284](https://github.com/bmadcode/BMAD-METHOD/issues/284)) ([1a4ca4f](https://github.com/bmadcode/BMAD-METHOD/commit/1a4ca4ffa630c2d4156bdd7a040d4c2274801757))
+
+## [4.22.1](https://github.com/bmadcode/BMAD-METHOD/compare/v4.22.0...v4.22.1) (2025-06-30)
+
+
+### Bug Fixes
+
+* update expansion versions ([6905fe7](https://github.com/bmadcode/BMAD-METHOD/commit/6905fe72f6c2abefbfd65729d1be85752130a1d2))
+
 # [4.22.0](https://github.com/bmadcode/BMAD-METHOD/compare/v4.21.2...v4.22.0) (2025-06-30)
 
 

README.md

@@ -156,7 +156,7 @@ The BMad Method works with any IDE, but has built-in integration for:
 - `gemini-cli` - Gemini with @agent commands
 - `windsurf` - Windsurf with manual rule @agent commands
 - `roo` - Roo Code with custom modes (see `.roomodes`)
-- `windsurf` - Windsurf with @agent commands
+- `vs-code-copilot` - VS Code with GitHub Copilot agent mode integration
 
 ## Available Agents
 
@@ -290,6 +290,7 @@ See the **[Core Architecture](docs/core-architecture.md)** for the complete sour
 - 🦘 [Roo Code Guide](docs/agentic-tools/roo-code-guide.md) - Setup and usage for Roo Code
 - 🔧 [Cline Guide](docs/agentic-tools/cline-guide.md) - Setup and usage for Cline (VS Code)
 - ✨ [Gemini CLI Guide](docs/agentic-tools/gemini-cli-guide.md) - Setup and usage for Gemini CLI
+- 💻 [VS Code Copilot Guide](docs/agentic-tools/vs-code-copilot-guide.md) - Setup and usage for VS Code with GitHub Copilot
 
 ## 🌟 Beyond Software Development - Expansion Packs
 

bmad-core/core-config.yml

@@ -1,4 +1,4 @@
-version: 4.20.1
+version: 4.21.1
 markdownExploder: true
 prd:
   prdFile: docs/prd.md

bmad-core/data/bmad-kb.md

@@ -78,7 +78,7 @@ BMAD transforms you into a "Vibe CEO" - directing a team of specialized AI agent
 5. Type `/help` to see available commands
 
 #### Option 2: IDE Integration
-**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code users
+**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code, VS Code Copilot users
 
 ```bash
 # Interactive installation (recommended)
@@ -93,6 +93,7 @@ npx bmad-method install
   - **Windsurf**: Built-in AI capabilities
   - **Cline**: VS Code extension with AI features
   - **Roo Code**: Web-based IDE with agent support
+  - **VS Code Copilot**: AI-powered coding assistant
 
 **Note for VS Code Users**: BMAD-METHOD assumes when you mention "VS Code" that you're using it with an AI-powered extension like GitHub Copilot, Cline, or Roo. Standard VS Code without AI capabilities cannot run BMAD agents. The installer includes built-in support for Cline and Roo.
 
@@ -278,6 +279,7 @@ You are the "Vibe CEO" - thinking like a CEO with unlimited resources and a sing
 - **Cursor**: `@agent-name` (e.g., `@bmad-master`)
 - **Windsurf**: `@agent-name` (e.g., `@bmad-master`)
 - **Roo Code**: Select mode from mode selector (e.g., `bmad-bmad-master`)
+- **VS Code Copilot**: Open the Chat view (`⌃⌘I` on Mac, `Ctrl+Alt+I` on Windows/Linux) and select **Agent** from the chat mode selector.
 
 **Chat Management Guidelines**:
 - **Claude Code, Cursor, Windsurf**: Start new chats when switching agents

dist/agents/analyst.txt

@@ -652,76 +652,73 @@ Present these numbered options to the user:
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#advanced-elicitation ====================
@@ -2013,7 +2010,7 @@ BMAD transforms you into a "Vibe CEO" - directing a team of specialized AI agent
 5. Type `/help` to see available commands
 
 #### Option 2: IDE Integration
-**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code users
+**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code, VS Code Copilot users
 
 ```bash
 # Interactive installation (recommended)

dist/agents/architect.txt

@@ -107,76 +107,73 @@ dependencies:
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#create-deep-research-prompt ====================

dist/agents/bmad-master.txt

@@ -1173,76 +1173,73 @@ Present these numbered options to the user:
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#document-project ====================
@@ -8399,7 +8396,7 @@ BMAD transforms you into a "Vibe CEO" - directing a team of specialized AI agent
 5. Type `/help` to see available commands
 
 #### Option 2: IDE Integration
-**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code users
+**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code, VS Code Copilot users
 
 ```bash
 # Interactive installation (recommended)

dist/agents/bmad-orchestrator.txt

@@ -263,76 +263,73 @@ Choose an action (0-9 - 9 to bypass - HELP for explanation of these options):
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#kb-mode-interaction ====================
@@ -489,7 +486,7 @@ BMAD transforms you into a "Vibe CEO" - directing a team of specialized AI agent
 5. Type `/help` to see available commands
 
 #### Option 2: IDE Integration
-**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code users
+**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code, VS Code Copilot users
 
 ```bash
 # Interactive installation (recommended)

dist/agents/pm.txt

@@ -104,76 +104,73 @@ dependencies:
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#correct-course ====================

dist/agents/ux-expert.txt

@@ -464,76 +464,73 @@ Present these numbered options to the user:
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#execute-checklist ====================

dist/expansion-packs/bmad-2d-phaser-game-dev/agents/game-designer.txt

@@ -104,76 +104,73 @@ dependencies:
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#execute-checklist ====================

dist/expansion-packs/bmad-2d-phaser-game-dev/teams/phaser-2d-nodejs-game-team.txt

@@ -977,76 +977,73 @@ Present these numbered options to the user:
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#advanced-elicitation ====================

dist/expansion-packs/bmad-infrastructure-devops/agents/infra-devops-platform.txt

@@ -105,76 +105,73 @@ dependencies:
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#review-infrastructure ====================

dist/teams/team-all.txt

@@ -755,76 +755,73 @@ Choose an action (0-9 - 9 to bypass - HELP for explanation of these options):
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#kb-mode-interaction ====================
@@ -981,7 +978,7 @@ BMAD transforms you into a "Vibe CEO" - directing a team of specialized AI agent
 5. Type `/help` to see available commands
 
 #### Option 2: IDE Integration
-**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code users
+**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code, VS Code Copilot users
 
 ```bash
 # Interactive installation (recommended)

dist/teams/team-fullstack.txt

@@ -599,76 +599,73 @@ Choose an action (0-9 - 9 to bypass - HELP for explanation of these options):
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#kb-mode-interaction ====================
@@ -825,7 +822,7 @@ BMAD transforms you into a "Vibe CEO" - directing a team of specialized AI agent
 5. Type `/help` to see available commands
 
 #### Option 2: IDE Integration
-**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code users
+**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code, VS Code Copilot users
 
 ```bash
 # Interactive installation (recommended)

dist/teams/team-ide-minimal.txt

@@ -499,76 +499,73 @@ Choose an action (0-9 - 9 to bypass - HELP for explanation of these options):
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#kb-mode-interaction ====================
@@ -725,7 +722,7 @@ BMAD transforms you into a "Vibe CEO" - directing a team of specialized AI agent
 5. Type `/help` to see available commands
 
 #### Option 2: IDE Integration
-**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code users
+**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code, VS Code Copilot users
 
 ```bash
 # Interactive installation (recommended)

dist/teams/team-no-ui.txt

@@ -531,76 +531,73 @@ Choose an action (0-9 - 9 to bypass - HELP for explanation of these options):
 
 ## Purpose
 
-- Generate documents from any specified template following embedded instructions from the perspective of the selected agent persona
+Generate documents from templates by EXECUTING (not just reading) embedded instructions from the perspective of the selected agent persona.
 
-## Instructions
+## CRITICAL RULES
 
-### 1. Identify Template and Context
+1. **Templates are PROGRAMS** - Execute every [[LLM:]] instruction exactly as written
+2. **NEVER show markup** - Hide all [[LLM:]], {{placeholders}}, @{examples}, and template syntax
+3. **STOP and EXECUTE** - When you see "apply tasks#" or "execute tasks#", STOP and run that task immediately
+4. **WAIT for user input** - At review points and after elicitation tasks
 
-- Determine which template to use (user-provided or list available for selection to user)
+## Execution Flow
 
-  - Agent-specific templates are listed in the agent's dependencies under `templates`. For each template listed, consider it a document the agent can create. So if an agent has:
+### 1. Identify Template
 
-    @{example}
+- Load from `templates#*` or `{root}/templates directory`
-    dependencies:
+- Agent-specific templates are listed in agent's dependencies
-    templates: - prd-tmpl - architecture-tmpl
+- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
-    @{/example}
 
-    You would offer to create "PRD" and "Architecture" documents when the user asks what you can help with.
+### 2. Ask Interaction Mode
 
-- Gather all relevant inputs, or ask for them, or else rely on user providing necessary details to complete the document
+> 1. **Incremental** - Section by section with reviews
-- Understand the document purpose and target audience
+> 2. **YOLO Mode** - Complete draft then review (user can type `/yolo` anytime to switch)
-
-### 2. Determine Interaction Mode
-
-Confirm with the user their preferred interaction style:
-
-- **Incremental:** Work through chunks of the document.
-- **YOLO Mode:** Draft complete document making reasonable assumptions in one shot. (Can be entered also after starting incremental by just typing /yolo)
 
 ### 3. Execute Template
 
-- Load specified template from `templates#*` or the `{root}/templates directory`
+- Replace {{placeholders}} with real content
-- Follow ALL embedded LLM instructions within the template
+- Execute [[LLM:]] instructions as you encounter them
-- Process template markup according to `utils#template-format` or `{root}/utils/template-format` conventions
+- Process <<REPEAT>> loops and ^^CONDITIONS^^
-
-### 4. Template Processing Rules
-
-#### CRITICAL: Never display template markup, LLM instructions, or examples to users
-
-- Replace all {{placeholders}} with actual content
-- Execute all [[LLM: instructions]] internally
-- Process `<<REPEAT>>` sections as needed
-- Evaluate ^^CONDITION^^ blocks and include only if applicable
 - Use @{examples} for guidance but never output them
 
-### 5. Content Generation
+### 4. Key Execution Patterns
 
-- **Incremental Mode**: Present each major section for review before proceeding
+**When you see:** `[[LLM: Draft X and immediately execute tasks#advanced-elicitation]]`
-- **YOLO Mode**: Generate all sections, then review complete document with user
-- Apply any elicitation protocols specified in template
-- Incorporate user feedback and iterate as needed
 
-### 6. Validation
+- Draft the content
+- Present it to user
+- IMMEDIATELY execute the task
+- Wait for completion before continuing
 
-If template specifies a checklist:
+**When you see:** `[[LLM: After section completion, apply tasks#Y]]`
 
-- Run the appropriate checklist against completed document
+- Finish the section
-- Document completion status for each item
+- STOP and execute the task
-- Address any deficiencies found
+- Wait for user input
-- Present validation summary to user
 
-### 7. Final Presentation
+### 5. Validation & Final Presentation
 
+- Run any specified checklists
 - Present clean, formatted content only
-- Ensure all sections are complete
+- No truncation or summarization
-- DO NOT truncate or summarize content
+- Begin directly with content (no preamble)
-- Begin directly with document content (no preamble)
+- Include any handoff prompts from template
-- Include any handoff prompts specified in template
 
-## Important Notes
+## Common Mistakes to Avoid
 
-- Template markup is for AI processing only - never expose to users
+❌ Skipping elicitation tasks
+❌ Showing template markup to users
+❌ Continuing past STOP signals
+❌ Combining multiple review points
+
+✅ Execute ALL instructions in sequence
+✅ Present only clean, formatted content
+✅ Stop at every elicitation point
+✅ Wait for user confirmation when instructed
+
+## Remember
+
+Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
 ==================== END: tasks#create-doc ====================
 
 ==================== START: tasks#kb-mode-interaction ====================
@@ -757,7 +754,7 @@ BMAD transforms you into a "Vibe CEO" - directing a team of specialized AI agent
 5. Type `/help` to see available commands
 
 #### Option 2: IDE Integration
-**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code users
+**Best for**: Cursor, Claude Code, Windsurf, Cline, Roo Code, VS Code Copilot users
 
 ```bash
 # Interactive installation (recommended)

docs/bmad-workflow-guide.md

@@ -22,7 +22,7 @@ The BMAD Method follows a structured approach to AI-assisted software developmen
    ```
 
    - Choose "Complete installation"
-   - Select your IDE (Cursor, Claude Code, Windsurf, or Roo Code)
+   - Select your IDE (Cursor, Claude Code, Windsurf, Roo Code, or VS Code Copilot)
 
 2. **Verify installation**:
    - `.bmad-core/` folder created with all agents
@@ -113,6 +113,7 @@ Follow the SM → Dev cycle for systematic story development:
 - **Cursor**: `@agent-name` (e.g., `@bmad-master`)
 - **Windsurf**: `@agent-name` (e.g., `@bmad-master`)
 - **Roo Code**: Select mode from mode selector (e.g., `bmad-bmad-master`)
+- **VS Code Copilot**: Open the Chat view (`⌃⌘I` on Mac, `Ctrl+Alt+I` on Windows/Linux) and select **Agent** from the chat mode selector.
 
 ### Chat Management:
 

docs/core-architecture.md

@@ -292,7 +292,8 @@ bmad-method/
 │   │   ├── cline-guide.md
 │   │   ├── gemini-cli-guide.md
 │   │   ├── roo-code-guide.md
-│   │   └── windsurf-guide.md
+│   │   ├── windsurf-guide.md
+│   │   └── vs-code-copilot-guide.md
 │   ├── bmad-workflow-guide.md    # Universal workflow guide
 │   ├── core-architecture.md      # This document
 │   ├── expansion-packs.md        # Expansion pack guide

expansion-packs/bmad-2d-phaser-game-dev/config.yml

@@ -1,5 +1,5 @@
 name: bmad-2d-phaser-game-dev
-version: 1.1.2
+version: 1.2.0
 short-title: 2D game development with Phaser 3 & TypeScript
 description: >-
   2D Game Development expansion pack for BMAD Method - Phaser 3 & TypeScript

expansion-packs/bmad-creator-tools/config.yml

@@ -1,5 +1,5 @@
 name: bmad-creator-tools
-version: 1.0.2
+version: 1.1.0
 short-title: Tools for creating BMAD framework components
 description: Tools for creating and extending BMAD framework components.
 author: Brian (BMad)

expansion-packs/bmad-infrastructure-devops/config.yml

@@ -1,5 +1,5 @@
 name: bmad-infrastructure-devops
-version: 1.0.2
+version: 1.1.0
 short-title: Infrastructure and DevOps capabilities
 description: >-
   This expansion pack extends BMAD Method with comprehensive infrastructure and

package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "bmad-method",
-  "version": "4.22.0",
+  "version": "4.23.0",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "bmad-method",
-      "version": "4.22.0",
+      "version": "4.23.0",
       "license": "MIT",
       "dependencies": {
         "@kayvan/markdown-tree-parser": "^1.5.0",

package.json

@@ -1,6 +1,6 @@
 {
   "name": "bmad-method",
-  "version": "4.22.0",
+  "version": "4.23.0",
   "description": "Breakthrough Method of Agile AI-driven Development",
   "main": "tools/cli.js",
   "bin": {

tools/installer/bin/bmad.js

@@ -50,7 +50,7 @@ program
   .option('-f, --full', 'Install complete BMAD Method')
   .option('-x, --expansion-only', 'Install only expansion packs (no bmad-core)')
   .option('-d, --directory <path>', 'Installation directory')
-  .option('-i, --ide <ide...>', 'Configure for specific IDE(s) - can specify multiple (cursor, claude-code, windsurf, roo, cline, gemini, other)')
+  .option('-i, --ide <ide...>', 'Configure for specific IDE(s) - can specify multiple (cursor, claude-code, windsurf, roo, cline, gemini, vs-code-copilot, other)')
   .option('-e, --expansion-packs <packs...>', 'Install specific expansion packs (can specify multiple)')
   .action(async (options) => {
     try {
@@ -236,7 +236,8 @@ async function promptInstallation() {
         { name: 'Windsurf', value: 'windsurf' },
         { name: 'Roo Code', value: 'roo' },
         { name: 'Cline', value: 'cline' },
-        { name: 'Gemini CLI', value: 'gemini' }
+        { name: 'Gemini CLI', value: 'gemini' },
+        { name: 'VS Code Copilot', value: 'vs-code-copilot' }
       ]
     }
   ]);

tools/installer/config/install.config.yml

@@ -67,3 +67,16 @@ ide-configurations:
       # 2. It also configures .gemini/settings.json to load all agent files.
       # 3. Simply mention the agent in your prompt (e.g., "As @dev, ...").
       # 4. The Gemini CLI will automatically have the context for that agent.
+  vs-code-copilot:
+    name: VS Code Copilot
+    rule-dir: .github/chatmodes/
+    format: multi-file
+    command-suffix: .md
+    instructions: |
+      # To use BMAD agents with VS Code Copilot:
+      # 1. The installer creates a .github/chatmodes/ directory in your project
+      # 2. Open the Chat view (`⌃⌘I` on Mac, `Ctrl+Alt+I` on Windows/Linux) and select **Agent** from the chat mode selector.
+      # 3. The agent will adopt that persona for the conversation
+      # 4. Requires VS Code 1.99+ with `chat.agent.enabled: true` in settings
+      # 5. Agent files are stored in .github/chatmodes/
+      # 6. Use `/help` to see available commands and agents

tools/installer/lib/ide-setup.js

@@ -6,12 +6,16 @@ const configLoader = require("./config-loader");
 
 // Dynamic import for ES module
 let chalk;
+let inquirer;
 
 // Initialize ES modules
 async function initializeModules() {
   if (!chalk) {
     chalk = (await import("chalk")).default;
   }
+  if (!inquirer) {
+    inquirer = (await import("inquirer")).default;
+  }
 }
 
 class IdeSetup {
@@ -36,7 +40,7 @@ class IdeSetup {
     }
   }
 
-  async setup(ide, installDir, selectedAgent = null) {
+  async setup(ide, installDir, selectedAgent = null, spinner = null) {
     await initializeModules();
     const ideConfig = await configLoader.getIdeConfiguration(ide);
 
@@ -58,6 +62,8 @@ class IdeSetup {
         return this.setupCline(installDir, selectedAgent);
       case "gemini":
         return this.setupGeminiCli(installDir, selectedAgent);
+      case "vs-code-copilot":
+        return this.setupVsCodeCopilot(installDir, selectedAgent, spinner);
       default:
         console.log(chalk.yellow(`\nIDE ${ide} not yet supported`));
         return false;
@@ -507,6 +513,213 @@ class IdeSetup {
 
     return true;
   }
+
+  async setupVsCodeCopilot(installDir, selectedAgent, spinner = null) {
+    await initializeModules();
+    
+    // Configure VS Code workspace settings first to avoid UI conflicts with loading spinners
+    await this.configureVsCodeSettings(installDir, spinner);
+    
+    const chatmodesDir = path.join(installDir, ".github", "chatmodes");
+    const agents = selectedAgent ? [selectedAgent] : await this.getAllAgentIds(installDir);
+     
+    await fileManager.ensureDirectory(chatmodesDir);
+
+    for (const agentId of agents) {
+      // Find the agent file
+      const agentPath = await this.findAgentPath(agentId, installDir);
+      const chatmodePath = path.join(chatmodesDir, `${agentId}.chatmode.md`);
+
+      if (agentPath) {
+        // Create chat mode file with agent content
+        const agentContent = await fileManager.readFile(agentPath);
+        const agentTitle = await this.getAgentTitle(agentId, installDir);
+        
+        // Extract whenToUse for the description
+        const yamlMatch = agentContent.match(/```ya?ml\n([\s\S]*?)```/);
+        let description = `Activates the ${agentTitle} agent persona.`;
+        if (yamlMatch) {
+          const whenToUseMatch = yamlMatch[1].match(/whenToUse:\s*"(.*?)"/);
+          if (whenToUseMatch && whenToUseMatch[1]) {
+            description = whenToUseMatch[1];
+          }
+        }
+        
+        let chatmodeContent = `---
+description: "${description.replace(/"/g, '\\"')}"
+tools: ['changes', 'codebase', 'fetch', 'findTestFiles', 'githubRepo', 'problems', 'usages']
+---
+
+`;
+        chatmodeContent += agentContent;
+
+        await fileManager.writeFile(chatmodePath, chatmodeContent);
+        console.log(chalk.green(`✓ Created chat mode: ${agentId}.chatmode.md`));
+      }
+    }
+
+    console.log(chalk.green(`\n✓ VS Code Copilot setup complete!`));
+    console.log(chalk.dim(`You can now find the BMAD agents in the Chat view's mode selector.`));
+
+    return true;
+  }
+
+  async configureVsCodeSettings(installDir, spinner) {
+    await initializeModules(); // Ensure inquirer is loaded
+    const vscodeDir = path.join(installDir, ".vscode");
+    const settingsPath = path.join(vscodeDir, "settings.json");
+    
+    await fileManager.ensureDirectory(vscodeDir);
+    
+    // Read existing settings if they exist
+    let existingSettings = {};
+    if (await fileManager.pathExists(settingsPath)) {
+      try {
+        const existingContent = await fileManager.readFile(settingsPath);
+        existingSettings = JSON.parse(existingContent);
+        console.log(chalk.yellow("Found existing .vscode/settings.json. Merging BMAD settings..."));
+      } catch (error) {
+        console.warn(chalk.yellow("Could not parse existing settings.json. Creating new one."));
+        existingSettings = {};
+      }
+    }
+    
+    // Clear any previous output and add spacing to avoid conflicts with loaders
+    console.log('\n'.repeat(2));
+    console.log(chalk.blue("🔧 VS Code Copilot Agent Settings Configuration"));
+    console.log(chalk.dim("BMAD works best with specific VS Code settings for optimal agent experience."));
+    console.log(''); // Add extra spacing
+    
+    const { configChoice } = await inquirer.prompt([
+      {
+        type: 'list',
+        name: 'configChoice',
+        message: 'How would you like to configure VS Code Copilot settings?',
+        choices: [
+          {
+            name: 'Use recommended defaults (fastest setup)',
+            value: 'defaults'
+          },
+          {
+            name: 'Configure each setting manually (customize to your preferences)',
+            value: 'manual'
+          },
+          {
+            name: 'Skip settings configuration (I\'ll configure manually later)\n',
+            value: 'skip'
+          }
+        ],
+        default: 'defaults'
+      }
+    ]);
+    
+    let bmadSettings = {};
+    
+    if (configChoice === 'skip') {
+      console.log(chalk.yellow("⚠️  Skipping VS Code settings configuration."));
+      console.log(chalk.dim("You can manually configure these settings in .vscode/settings.json:"));
+      console.log(chalk.dim("  • chat.agent.enabled: true"));
+      console.log(chalk.dim("  • chat.agent.maxRequests: 15"));
+      console.log(chalk.dim("  • github.copilot.chat.agent.runTasks: true"));
+      console.log(chalk.dim("  • chat.mcp.discovery.enabled: true"));
+      console.log(chalk.dim("  • github.copilot.chat.agent.autoFix: true"));
+      console.log(chalk.dim("  • chat.tools.autoApprove: false"));
+      return true;
+    }
+    
+    if (configChoice === 'defaults') {
+      // Use recommended defaults
+      bmadSettings = {
+        "chat.agent.enabled": true,
+        "chat.agent.maxRequests": 15,
+        "github.copilot.chat.agent.runTasks": true,
+        "chat.mcp.discovery.enabled": true,
+        "github.copilot.chat.agent.autoFix": true,
+        "chat.tools.autoApprove": false
+      };
+      console.log(chalk.green("✓ Using recommended BMAD defaults for VS Code Copilot settings"));
+    } else {
+      // Manual configuration
+      console.log(chalk.blue("\n📋 Let's configure each setting for your preferences:"));
+      
+      // Pause spinner during manual configuration prompts
+      let spinnerWasActive = false;
+      if (spinner && spinner.isSpinning) {
+        spinner.stop();
+        spinnerWasActive = true;
+      }
+      
+      const manualSettings = await inquirer.prompt([
+        {
+          type: 'input',
+          name: 'maxRequests',
+          message: 'Maximum requests per agent session (recommended: 15)?',
+          default: '15',
+          validate: (input) => {
+            const num = parseInt(input);
+            if (isNaN(num) || num < 1 || num > 50) {
+              return 'Please enter a number between 1 and 50';
+            }
+            return true;
+          }
+        },
+        {
+          type: 'confirm',
+          name: 'runTasks',
+          message: 'Allow agents to run workspace tasks (package.json scripts, etc.)?',
+          default: true
+        },
+        {
+          type: 'confirm',
+          name: 'mcpDiscovery',
+          message: 'Enable MCP (Model Context Protocol) server discovery?',
+          default: true
+        },
+        {
+          type: 'confirm',
+          name: 'autoFix',
+          message: 'Enable automatic error detection and fixing in generated code?',
+          default: true
+        },
+        {
+          type: 'confirm',
+          name: 'autoApprove',
+          message: 'Auto-approve ALL tools without confirmation? (⚠️  EXPERIMENTAL - less secure)',
+          default: false
+        }
+      ]);
+
+      // Restart spinner if it was active before prompts
+      if (spinner && spinnerWasActive) {
+        spinner.start();
+      }
+      
+      bmadSettings = {
+        "chat.agent.enabled": true, // Always enabled - required for BMAD agents
+        "chat.agent.maxRequests": parseInt(manualSettings.maxRequests),
+        "github.copilot.chat.agent.runTasks": manualSettings.runTasks,
+        "chat.mcp.discovery.enabled": manualSettings.mcpDiscovery,
+        "github.copilot.chat.agent.autoFix": manualSettings.autoFix,
+        "chat.tools.autoApprove": manualSettings.autoApprove
+      };
+      
+      console.log(chalk.green("✓ Custom settings configured"));
+    }
+    
+    // Merge settings (existing settings take precedence to avoid overriding user preferences)
+    const mergedSettings = { ...bmadSettings, ...existingSettings };
+    
+    // Write the updated settings
+    await fileManager.writeFile(settingsPath, JSON.stringify(mergedSettings, null, 2));
+    
+    console.log(chalk.green("✓ VS Code workspace settings configured successfully"));
+    console.log(chalk.dim("  Settings written to .vscode/settings.json:"));
+    Object.entries(bmadSettings).forEach(([key, value]) => {
+      console.log(chalk.dim(`  • ${key}: ${value}`));
+    });
+    console.log(chalk.dim(""));
+    console.log(chalk.dim("You can modify these settings anytime in .vscode/settings.json"));
+  }
 }
 
 module.exports = new IdeSetup();

tools/installer/lib/installer.js

@@ -372,7 +372,7 @@ class Installer {
     if (ides.length > 0) {
       for (const ide of ides) {
         spinner.text = `Setting up ${ide} integration...`;
-        await ideSetup.setup(ide, installDir, config.agent);
+        await ideSetup.setup(ide, installDir, config.agent, spinner);
       }
     }
 

tools/installer/package.json

@@ -1,6 +1,6 @@
 {
   "name": "bmad-method",
-  "version": "4.22.0",
+  "version": "4.23.0",
   "description": "BMAD Method installer - AI-powered Agile development framework",
   "main": "lib/installer.js",
   "bin": {