ros/BMAD-METHOD
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: corrected cursor agent update instructions

Browse Source
This commit is contained in:
Brian Madison
2025-07-01 22:54:39 -05:00
parent b89aa48f7b
commit 84e394ac11
14 changed files with 5289 additions and 673 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
dist/agents/analyst.txt vendored
View File

@@ -670,7 +670,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- If `workflow.trackProgress: true`, check for active plan using utils#plan-management
- If plan exists and this document creation is part of the plan:
- Verify this is the expected next step
- If out of sequence and `enforceSequence: true`, warn user
- If out of sequence and `enforceSequence: true`, warn user and halt without user override
- If out of sequence and `enforceSequence: false`, ask for confirmation
- Continue with normal execution after plan check
@@ -678,7 +678,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- Load from `templates#*` or `{root}/templates directory`
- Agent-specific templates are listed in agent's dependencies
- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
- If agent has `templates: [prd-tmpl, architecture-tmpl]` for example, then offer to create "PRD" and "Architecture" documents
### 2. Ask Interaction Mode
@@ -2045,6 +2045,7 @@ 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.
@@ -2230,6 +2231,7 @@ 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

4
dist/agents/architect.txt vendored
View File

@@ -125,7 +125,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- If `workflow.trackProgress: true`, check for active plan using utils#plan-management
- If plan exists and this document creation is part of the plan:
- Verify this is the expected next step
- If out of sequence and `enforceSequence: true`, warn user
- If out of sequence and `enforceSequence: true`, warn user and halt without user override
- If out of sequence and `enforceSequence: false`, ask for confirmation
- Continue with normal execution after plan check
@@ -133,7 +133,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- Load from `templates#*` or `{root}/templates directory`
- Agent-specific templates are listed in agent's dependencies
- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
- If agent has `templates: [prd-tmpl, architecture-tmpl]` for example, then offer to create "PRD" and "Architecture" documents
### 2. Ask Interaction Mode

14
dist/agents/bmad-master.txt vendored
View File

@@ -132,7 +132,6 @@ dependencies:
- bmad-kb
- technical-preferences
utils:
- agent-switcher.ide
- plan-management
- template-format
- workflow-management
@@ -1205,7 +1204,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- If `workflow.trackProgress: true`, check for active plan using utils#plan-management
- If plan exists and this document creation is part of the plan:
- Verify this is the expected next step
- If out of sequence and `enforceSequence: true`, warn user
- If out of sequence and `enforceSequence: true`, warn user and halt without user override
- If out of sequence and `enforceSequence: false`, ask for confirmation
- Continue with normal execution after plan check
@@ -1213,7 +1212,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- Load from `templates#*` or `{root}/templates directory`
- Agent-specific templates are listed in agent's dependencies
- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
- If agent has `templates: [prd-tmpl, architecture-tmpl]` for example, then offer to create "PRD" and "Architecture" documents
### 2. Ask Interaction Mode
@@ -2708,6 +2707,7 @@ Update the status of steps in an active workflow plan, mark completions, add not
[[LLM: First load core-config.yml to get plan settings]]
Check workflow configuration:
- `workflow.planFile` - Location of the plan (default: docs/workflow-plan.md)
- `workflow.trackProgress` - Whether tracking is enabled
- `workflow.updateOnCompletion` - Whether to auto-update on task completion
@@ -2719,6 +2719,7 @@ If tracking is disabled, inform user and exit.
[[LLM: Check if workflow plan exists at configured location]]
If no plan exists:
```
No active workflow plan found at {location}.
Would you like to create one? Use *plan command.
@@ -2748,6 +2749,7 @@ Please select an option (1-6):
[[LLM: Read and parse the plan to understand current state]]
Extract:
- All steps with their checkbox status
- Step IDs from comments (if present)
- Current completion percentage
@@ -2814,6 +2816,7 @@ If user selected option 5:
[[LLM: When called automatically by another task]]
If called with parameters:
```
task: {task_name}
step_id: {step_identifier}
@@ -2822,6 +2825,7 @@ note: {optional_note}
```
Automatically:
1. Find the corresponding step
2. Update its status
3. Add completion metadata
@@ -2854,6 +2858,7 @@ Plan location: {file_path}
Other tasks can integrate by:
1. **After Task Completion**:
```
At end of task execution:
- Check if task corresponds to a plan step
@@ -2864,6 +2869,7 @@ At end of task execution:
```
2. **On Task Failure**:
```
If task fails:
- Call update-workflow-plan with:
@@ -9030,6 +9036,7 @@ 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.
@@ -9215,6 +9222,7 @@ 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

13
dist/agents/bmad-orchestrator.txt vendored
View File

@@ -295,7 +295,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- If `workflow.trackProgress: true`, check for active plan using utils#plan-management
- If plan exists and this document creation is part of the plan:
- Verify this is the expected next step
- If out of sequence and `enforceSequence: true`, warn user
- If out of sequence and `enforceSequence: true`, warn user and halt without user override
- If out of sequence and `enforceSequence: false`, ask for confirmation
- Continue with normal execution after plan check
@@ -303,7 +303,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- Load from `templates#*` or `{root}/templates directory`
- Agent-specific templates are listed in agent's dependencies
- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
- If agent has `templates: [prd-tmpl, architecture-tmpl]` for example, then offer to create "PRD" and "Architecture" documents
### 2. Ask Interaction Mode
@@ -745,6 +745,7 @@ Update the status of steps in an active workflow plan, mark completions, add not
[[LLM: First load core-config.yml to get plan settings]]
Check workflow configuration:
- `workflow.planFile` - Location of the plan (default: docs/workflow-plan.md)
- `workflow.trackProgress` - Whether tracking is enabled
- `workflow.updateOnCompletion` - Whether to auto-update on task completion
@@ -756,6 +757,7 @@ If tracking is disabled, inform user and exit.
[[LLM: Check if workflow plan exists at configured location]]
If no plan exists:
```
No active workflow plan found at {location}.
Would you like to create one? Use *plan command.
@@ -785,6 +787,7 @@ Please select an option (1-6):
[[LLM: Read and parse the plan to understand current state]]
Extract:
- All steps with their checkbox status
- Step IDs from comments (if present)
- Current completion percentage
@@ -851,6 +854,7 @@ If user selected option 5:
[[LLM: When called automatically by another task]]
If called with parameters:
```
task: {task_name}
step_id: {step_identifier}
@@ -859,6 +863,7 @@ note: {optional_note}
```
Automatically:
1. Find the corresponding step
2. Update its status
3. Add completion metadata
@@ -891,6 +896,7 @@ Plan location: {file_path}
Other tasks can integrate by:
1. **After Task Completion**:
```
At end of task execution:
- Check if task corresponds to a plan step
@@ -901,6 +907,7 @@ At end of task execution:
```
2. **On Task Failure**:
```
If task fails:
- Call update-workflow-plan with:
@@ -1071,6 +1078,7 @@ 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.
@@ -1256,6 +1264,7 @@ 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

4
dist/agents/pm.txt vendored
View File

@@ -122,7 +122,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- If `workflow.trackProgress: true`, check for active plan using utils#plan-management
- If plan exists and this document creation is part of the plan:
- Verify this is the expected next step
- If out of sequence and `enforceSequence: true`, warn user
- If out of sequence and `enforceSequence: true`, warn user and halt without user override
- If out of sequence and `enforceSequence: false`, ask for confirmation
- Continue with normal execution after plan check
@@ -130,7 +130,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- Load from `templates#*` or `{root}/templates directory`
- Agent-specific templates are listed in agent's dependencies
- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
- If agent has `templates: [prd-tmpl, architecture-tmpl]` for example, then offer to create "PRD" and "Architecture" documents
### 2. Ask Interaction Mode

4
dist/agents/ux-expert.txt vendored
View File

@@ -482,7 +482,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- If `workflow.trackProgress: true`, check for active plan using utils#plan-management
- If plan exists and this document creation is part of the plan:
- Verify this is the expected next step
- If out of sequence and `enforceSequence: true`, warn user
- If out of sequence and `enforceSequence: true`, warn user and halt without user override
- If out of sequence and `enforceSequence: false`, ask for confirmation
- Continue with normal execution after plan check
@@ -490,7 +490,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- Load from `templates#*` or `{root}/templates directory`
- Agent-specific templates are listed in agent's dependencies
- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
- If agent has `templates: [prd-tmpl, architecture-tmpl]` for example, then offer to create "PRD" and "Architecture" documents
### 2. Ask Interaction Mode

4
dist/expansion-packs/bmad-2d-phaser-game-dev/agents/game-designer.txt vendored
View File

@@ -122,7 +122,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- If `workflow.trackProgress: true`, check for active plan using utils#plan-management
- If plan exists and this document creation is part of the plan:
- Verify this is the expected next step
- If out of sequence and `enforceSequence: true`, warn user
- If out of sequence and `enforceSequence: true`, warn user and halt without user override
- If out of sequence and `enforceSequence: false`, ask for confirmation
- Continue with normal execution after plan check
@@ -130,7 +130,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- Load from `templates#*` or `{root}/templates directory`
- Agent-specific templates are listed in agent's dependencies
- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
- If agent has `templates: [prd-tmpl, architecture-tmpl]` for example, then offer to create "PRD" and "Architecture" documents
### 2. Ask Interaction Mode

11
dist/expansion-packs/bmad-2d-phaser-game-dev/teams/phaser-2d-nodejs-game-team.txt vendored
View File

@@ -1009,7 +1009,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- If `workflow.trackProgress: true`, check for active plan using utils#plan-management
- If plan exists and this document creation is part of the plan:
- Verify this is the expected next step
- If out of sequence and `enforceSequence: true`, warn user
- If out of sequence and `enforceSequence: true`, warn user and halt without user override
- If out of sequence and `enforceSequence: false`, ask for confirmation
- Continue with normal execution after plan check
@@ -1017,7 +1017,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- Load from `templates#*` or `{root}/templates directory`
- Agent-specific templates are listed in agent's dependencies
- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
- If agent has `templates: [prd-tmpl, architecture-tmpl]` for example, then offer to create "PRD" and "Architecture" documents
### 2. Ask Interaction Mode
@@ -2972,6 +2972,7 @@ Update the status of steps in an active workflow plan, mark completions, add not
[[LLM: First load core-config.yml to get plan settings]]
Check workflow configuration:
- `workflow.planFile` - Location of the plan (default: docs/workflow-plan.md)
- `workflow.trackProgress` - Whether tracking is enabled
- `workflow.updateOnCompletion` - Whether to auto-update on task completion
@@ -2983,6 +2984,7 @@ If tracking is disabled, inform user and exit.
[[LLM: Check if workflow plan exists at configured location]]
If no plan exists:
```
No active workflow plan found at {location}.
Would you like to create one? Use *plan command.
@@ -3012,6 +3014,7 @@ Please select an option (1-6):
[[LLM: Read and parse the plan to understand current state]]
Extract:
- All steps with their checkbox status
- Step IDs from comments (if present)
- Current completion percentage
@@ -3078,6 +3081,7 @@ If user selected option 5:
[[LLM: When called automatically by another task]]
If called with parameters:
```
task: {task_name}
step_id: {step_identifier}
@@ -3086,6 +3090,7 @@ note: {optional_note}
```
Automatically:
1. Find the corresponding step
2. Update its status
3. Add completion metadata
@@ -3118,6 +3123,7 @@ Plan location: {file_path}
Other tasks can integrate by:
1. **After Task Completion**:
```
At end of task execution:
- Check if task corresponds to a plan step
@@ -3128,6 +3134,7 @@ At end of task execution:
```
2. **On Task Failure**:
```
If task fails:
- Call update-workflow-plan with:

4
dist/expansion-packs/bmad-infrastructure-devops/agents/infra-devops-platform.txt vendored
View File

@@ -123,7 +123,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- If `workflow.trackProgress: true`, check for active plan using utils#plan-management
- If plan exists and this document creation is part of the plan:
- Verify this is the expected next step
- If out of sequence and `enforceSequence: true`, warn user
- If out of sequence and `enforceSequence: true`, warn user and halt without user override
- If out of sequence and `enforceSequence: false`, ask for confirmation
- Continue with normal execution after plan check
@@ -131,7 +131,7 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- Load from `templates#*` or `{root}/templates directory`
- Agent-specific templates are listed in agent's dependencies
- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
- If agent has `templates: [prd-tmpl, architecture-tmpl]` for example, then offer to create "PRD" and "Architecture" documents
### 2. Ask Interaction Mode

1947
dist/teams/team-all.txt vendored
View File

File diff suppressed because it is too large Load Diff

1922
dist/teams/team-fullstack.txt vendored
View File

File diff suppressed because it is too large Load Diff

832
dist/teams/team-ide-minimal.txt vendored
View File

@@ -82,6 +82,9 @@ startup:
- Announce: Introduce yourself as the BMAD Orchestrator, explain you can coordinate agents and workflows
- IMPORTANT: Tell users that all commands start with * (e.g., *help, *agent, *workflow)
- Mention *help shows all available commands and options
- Check for active workflow plan using utils#plan-management
- 'If plan exists: Show 📋 Active plan: {workflow} ({progress}% complete). Use *plan-status for details.'
- 'If plan exists: Suggest next action based on plan progress'
- Assess user goal against available agents and workflows in this bundle
- If clear match to an agent's expertise, suggest transformation with *agent command
- If project-oriented, suggest *workflow-guidance to explore options
@@ -96,6 +99,9 @@ commands:
task: Run a specific task (list if name not specified)
workflow: Start a specific workflow (list if name not specified)
workflow-guidance: Get personalized help selecting the right workflow
plan: Create detailed workflow plan before starting
plan-status: Show current workflow plan progress
plan-update: Update workflow plan status
checklist: Execute a checklist (list if name not specified)
yolo: Toggle skip confirmations mode
party-mode: Group chat with all agents
@@ -119,6 +125,9 @@ help-display-template: |
Workflow Commands:
*workflow [name] .... Start specific workflow (list if no name)
*workflow-guidance .. Get personalized help selecting the right workflow
*plan ............... Create detailed workflow plan before starting
*plan-status ........ Show current workflow plan progress
*plan-update ........ Update workflow plan status
Other Commands:
*yolo ............... Toggle skip confirmations mode
@@ -159,6 +168,8 @@ workflow-guidance:
- Understand each workflow's purpose, options, and decision points
- Ask clarifying questions based on the workflow's structure
- Guide users through workflow selection when multiple options exist
- For complex projects, offer to create a workflow plan using create-workflow-plan task
- When appropriate, suggest: Would you like me to create a detailed workflow plan before starting?
- For workflows with divergent paths, help users choose the right path
- Adapt questions to the specific domain (e.g., game dev vs infrastructure vs web dev)
- Only recommend workflows that actually exist in the current bundle
@@ -167,10 +178,13 @@ dependencies:
tasks:
- advanced-elicitation
- create-doc
- create-workflow-plan
- kb-mode-interaction
- update-workflow-plan
data:
- bmad-kb
utils:
- plan-management
- workflow-management
- template-format
```
@@ -510,11 +524,22 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
## Execution Flow
### 0. Check Workflow Plan (if configured)
[[LLM: Check if plan tracking is enabled in core-config.yml]]
- If `workflow.trackProgress: true`, check for active plan using utils#plan-management
- If plan exists and this document creation is part of the plan:
- Verify this is the expected next step
- If out of sequence and `enforceSequence: true`, warn user and halt without user override
- If out of sequence and `enforceSequence: false`, ask for confirmation
- Continue with normal execution after plan check
### 1. Identify Template
- Load from `templates#*` or `{root}/templates directory`
- Agent-specific templates are listed in agent's dependencies
- If agent has `templates: [prd-tmpl, architecture-tmpl]`, offer to create "PRD" and "Architecture" documents
- If agent has `templates: [prd-tmpl, architecture-tmpl]` for example, then offer to create "PRD" and "Architecture" documents
### 2. Ask Interaction Mode
@@ -551,6 +576,15 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
- Begin directly with content (no preamble)
- Include any handoff prompts from template
### 6. Update Workflow Plan (if applicable)
[[LLM: After successful document creation]]
- If plan tracking is enabled and document was part of plan:
- Call update-workflow-plan task to mark step complete
- Parameters: task: create-doc, step_id: {from plan}, status: complete
- Show next recommended step from plan
## Common Mistakes to Avoid
❌ Skipping elicitation tasks
@@ -568,6 +602,298 @@ Generate documents from templates by EXECUTING (not just reading) embedded instr
Templates contain precise instructions for a reason. Follow them exactly to ensure document quality and completeness.
==================== END: tasks#create-doc ====================
==================== START: tasks#create-workflow-plan ====================
# Create Workflow Plan Task
## Purpose
Guide users through workflow selection and create a detailed plan document that outlines the selected workflow steps, decision points, and expected outputs. This task helps users understand what will happen before starting a complex workflow and provides a checklist to track progress.
## Task Instructions
### 1. Understand User's Goal
[[LLM: Start with discovery questions to understand what the user wants to accomplish]]
Ask the user:
1. **Project Type**:
- Are you starting a new project (greenfield) or enhancing an existing one (brownfield)?
- What type of application? (web app, service/API, UI only, full-stack)
2. **For Greenfield**:
- Do you need a quick prototype or production-ready application?
- Will this have a UI component?
- Single service or multiple services?
3. **For Brownfield**:
- What's the scope of the enhancement?
- Single bug fix or small feature (few hours)
- Small enhancement (1-3 stories)
- Major feature requiring coordination
- Architectural changes or modernization
- Do you have existing documentation?
- Are you following existing patterns or introducing new ones?
### 2. Recommend Appropriate Workflow
Based on the answers, recommend:
**Greenfield Options:**
- `greenfield-fullstack` - Complete web application
- `greenfield-service` - Backend API/service only
- `greenfield-ui` - Frontend only
**Brownfield Options:**
- `brownfield-create-story` - Single small change
- `brownfield-create-epic` - Small feature (1-3 stories)
- `brownfield-fullstack` - Major enhancement
**Simplified Option:**
- For users unsure or wanting flexibility, suggest starting with individual agent tasks
### 3. Explain Selected Workflow
[[LLM: Once workflow is selected, provide clear explanation]]
For the selected workflow, explain:
1. **Overview**: What this workflow accomplishes
2. **Duration**: Estimated time for planning phase
3. **Outputs**: What documents will be created
4. **Decision Points**: Where user input will be needed
5. **Requirements**: What information should be ready
### 4. Create Workflow Plan Document
[[LLM: Generate a comprehensive plan document with the following structure]]
```markdown
# Workflow Plan: {{Workflow Name}}
<!-- WORKFLOW-PLAN-META
workflow-id: {{workflow-id}}
status: active
created: {{ISO-8601 timestamp}}
updated: {{ISO-8601 timestamp}}
version: 1.0
-->
**Created Date**: {{current date}}
**Project**: {{project name}}
**Type**: {{greenfield/brownfield}}
**Status**: Active
**Estimated Planning Duration**: {{time estimate}}
## Objective
{{Clear description of what will be accomplished}}
## Selected Workflow
**Workflow**: `{{workflow-id}}`
**Reason**: {{Why this workflow fits the user's needs}}
## Workflow Steps
### Planning Phase
- [ ] Step 1: {{step name}} <!-- step-id: 1.1, agent: {{agent}}, task: {{task}} -->
- **Agent**: {{agent name}}
- **Action**: {{what happens}}
- **Output**: {{what's created}}
- **User Input**: {{if any}}
- [ ] Step 2: {{step name}} <!-- step-id: 1.2, agent: {{agent}}, task: {{task}} -->
- **Agent**: {{agent name}}
- **Action**: {{what happens}}
- **Output**: {{what's created}}
- **Decision Point**: {{if any}} <!-- decision-id: D1 -->
{{Continue for all planning steps}}
### Development Phase (IDE)
- [ ] Document Sharding <!-- step-id: 2.1, agent: po, task: shard-doc -->
- Prepare documents for story creation
- [ ] Story Development Cycle <!-- step-id: 2.2, repeats: true -->
- [ ] Create story (SM agent) <!-- step-id: 2.2.1, agent: sm, task: create-next-story -->
- [ ] Review story (optional) <!-- step-id: 2.2.2, agent: analyst, optional: true -->
- [ ] Implement story (Dev agent) <!-- step-id: 2.2.3, agent: dev -->
- [ ] QA review (optional) <!-- step-id: 2.2.4, agent: qa, optional: true -->
- [ ] Repeat for all stories
- [ ] Epic Retrospective (optional) <!-- step-id: 2.3, agent: po, optional: true -->
## Key Decision Points
1. **{{Decision Name}}** (Step {{n}}): <!-- decision-id: D1, status: pending -->
- Trigger: {{what causes this decision}}
- Options: {{available choices}}
- Impact: {{how it affects the workflow}}
- Decision Made: _Pending_
{{List all decision points}}
## Expected Outputs
### Planning Documents
- [ ] {{document 1}} - {{description}}
- [ ] {{document 2}} - {{description}}
{{etc...}}
### Development Artifacts
- [ ] Stories in `docs/stories/`
- [ ] Implementation code
- [ ] Tests
- [ ] Updated documentation
## Prerequisites Checklist
Before starting this workflow, ensure you have:
- [ ] {{prerequisite 1}}
- [ ] {{prerequisite 2}}
- [ ] {{prerequisite 3}}
{{etc...}}
## Customization Options
Based on your project needs, you may:
- Skip {{optional step}} if {{condition}}
- Add {{additional step}} if {{condition}}
- Choose {{alternative}} instead of {{default}}
## Risk Considerations
{{For brownfield only}}
- Integration complexity: {{assessment}}
- Rollback strategy: {{approach}}
- Testing requirements: {{special needs}}
## Next Steps
1. Review this plan and confirm it matches your expectations
2. Gather any missing prerequisites
3. Start workflow with: `*task workflow {{workflow-id}}`
4. Or begin with first agent: `@{{first-agent}}`
## Notes
{{Any additional context or warnings}}
---
*This plan can be updated as you progress through the workflow. Check off completed items to track progress.*
```
### 5. Save and Present Plan
1. Save the plan as `docs/workflow-plan.md`
2. Inform user: "Workflow plan created at docs/workflow-plan.md"
3. Offer options:
- Review the plan together
- Start the workflow now
- Gather prerequisites first
- Modify the plan
### 6. Plan Variations
[[LLM: Adjust plan detail based on workflow complexity]]
**For Simple Workflows** (create-story, create-epic):
- Simpler checklist format
- Focus on immediate next steps
- Less detailed explanations
**For Complex Workflows** (full greenfield/brownfield):
- Detailed step breakdowns
- All decision points documented
- Comprehensive output descriptions
- Risk mitigation sections
**For Brownfield Workflows**:
- Include existing system impact analysis
- Document integration checkpoints
- Add rollback considerations
- Note documentation dependencies
### 7. Interactive Planning Mode
[[LLM: If user wants to customize the workflow]]
If user wants to modify the standard workflow:
1. Present workflow steps as options
2. Allow skipping optional steps
3. Let user reorder certain steps
4. Document customizations in plan
5. Warn about dependencies if steps are skipped
### 8. Execution Guidance
After plan is created, provide clear guidance:
```text
Your workflow plan is ready! Here's how to proceed:
1. **Review the plan**: Check that all steps align with your goals
2. **Gather prerequisites**: Use the checklist to ensure you're ready
3. **Start execution**:
- Full workflow: `*task workflow {{workflow-id}}`
- Step by step: Start with `@{{first-agent}}`
4. **Track progress**: Check off steps in the plan as completed
Would you like to:
a) Review the plan together
b) Start the workflow now
c) Gather prerequisites first
d) Modify the plan
```
## Success Criteria
The workflow plan is successful when:
1. User clearly understands what will happen
2. All decision points are documented
3. Prerequisites are identified
4. Expected outputs are clear
5. User feels confident to proceed
6. Plan serves as useful progress tracker
## Integration with BMad Master and Orchestrator
When used by BMad Master or BMad Orchestrator, this task should:
1. Be offered when user asks about workflows
2. Be suggested before starting complex workflows
3. Create a plan that the agent can reference during execution
4. Allow the agent to track progress against the plan
## Example Usage
```text
User: "I need to add a payment system to my existing app"
BMad Orchestrator: "Let me help you create a workflow plan for that enhancement. I'll ask a few questions to recommend the best approach..."
[Runs through discovery questions]
BMad Orchestrator: "Based on your answers, I recommend the brownfield-fullstack workflow. Let me create a detailed plan for you..."
[Creates and saves plan]
BMad Orchestrator: "I've created a workflow plan at docs/workflow-plan.md. This shows all the steps we'll go through, what documents will be created, and where you'll need to make decisions. Would you like to review it together?"
```
==================== END: tasks#create-workflow-plan ====================
==================== START: tasks#kb-mode-interaction ====================
# KB Mode Interaction Task
@@ -641,6 +967,257 @@ Or ask me about anything else related to BMAD-METHOD!
**Assistant**: [Provides focused information about workflows from the KB, then offers to explore specific workflow types or related topics]
==================== END: tasks#kb-mode-interaction ====================
==================== START: tasks#update-workflow-plan ====================
# Update Workflow Plan Task
## Purpose
Update the status of steps in an active workflow plan, mark completions, add notes about deviations, and maintain an accurate record of workflow progress. This task can be called directly by users or automatically by other tasks upon completion.
## Task Instructions
### 0. Load Plan Configuration
[[LLM: First load core-config.yml to get plan settings]]
Check workflow configuration:
- `workflow.planFile` - Location of the plan (default: docs/workflow-plan.md)
- `workflow.trackProgress` - Whether tracking is enabled
- `workflow.updateOnCompletion` - Whether to auto-update on task completion
If tracking is disabled, inform user and exit.
### 1. Verify Plan Exists
[[LLM: Check if workflow plan exists at configured location]]
If no plan exists:
```
No active workflow plan found at {location}.
Would you like to create one? Use *plan command.
```
### 2. Determine Update Type
[[LLM: Ask user what type of update they want to make]]
Present options:
```
What would you like to update in the workflow plan?
1. Mark step as complete
2. Update current step
3. Add deviation note
4. Mark decision point resolution
5. Update overall status
6. View current plan status only
Please select an option (1-6):
```
### 3. Parse Current Plan
[[LLM: Read and parse the plan to understand current state]]
Extract:
- All steps with their checkbox status
- Step IDs from comments (if present)
- Current completion percentage
- Any existing deviation notes
- Decision points and their status
### 4. Execute Updates
#### 4.1 Mark Step Complete
If user selected option 1:
1. Show numbered list of incomplete steps
2. Ask which step to mark complete
3. Update the checkbox from `[ ]` to `[x]`
4. Add completion timestamp: `<!-- completed: YYYY-MM-DD HH:MM -->`
5. If this was the current step, identify next step
#### 4.2 Update Current Step
If user selected option 2:
1. Show all steps with current status
2. Ask which step is now current
3. Add/move `<!-- current-step -->` marker
4. Optionally add note about why sequence changed
#### 4.3 Add Deviation Note
If user selected option 3:
1. Ask for deviation description
2. Ask which step this relates to (or general)
3. Insert note in appropriate location:
```markdown
> **Deviation Note** (YYYY-MM-DD): {user_note}
> Related to: Step X.Y or General workflow
```
#### 4.4 Mark Decision Resolution
If user selected option 4:
1. Show pending decision points
2. Ask which decision was made
3. Record the decision and chosen path
4. Update related steps based on decision
#### 4.5 Update Overall Status
If user selected option 5:
1. Show current overall status
2. Provide options:
- Active (continuing with plan)
- Paused (temporarily stopped)
- Abandoned (no longer following)
- Complete (all steps done)
3. Update plan header with new status
### 5. Automatic Updates (When Called by Tasks)
[[LLM: When called automatically by another task]]
If called with parameters:
```
task: {task_name}
step_id: {step_identifier}
status: complete|skipped|failed
note: {optional_note}
```
Automatically:
1. Find the corresponding step
2. Update its status
3. Add completion metadata
4. Add note if provided
5. Calculate new progress percentage
### 6. Generate Update Summary
After updates, show summary:
```
✅ Workflow Plan Updated
Changes made:
- {change_1}
- {change_2}
New Status:
- Progress: {X}% complete ({completed}/{total} steps)
- Current Step: {current_step}
- Next Recommended: {next_step}
Plan location: {file_path}
```
### 7. Integration with Other Tasks
[[LLM: How other tasks should call this]]
Other tasks can integrate by:
1. **After Task Completion**:
```
At end of task execution:
- Check if task corresponds to a plan step
- If yes, call update-workflow-plan with:
- task: {current_task_name}
- step_id: {matching_step}
- status: complete
```
2. **On Task Failure**:
```
If task fails:
- Call update-workflow-plan with:
- task: {current_task_name}
- status: failed
- note: {failure_reason}
```
### 8. Plan Status Display
[[LLM: When user selects view status only]]
Display comprehensive status:
```markdown
📋 Workflow Plan Status
━━━━━━━━━━━━━━━━━━━━
Workflow: {workflow_name}
Status: {Active|Paused|Complete}
Progress: {X}% complete ({completed}/{total} steps)
Last Updated: {timestamp}
✅ Completed Steps:
- [x] Step 1.1: {description} (completed: {date})
- [x] Step 1.2: {description} (completed: {date})
🔄 Current Step:
- [ ] Step 2.1: {description} <!-- current-step -->
Agent: {agent_name}
Task: {task_name}
📌 Upcoming Steps:
- [ ] Step 2.2: {description}
- [ ] Step 3.1: {description}
⚠️ Deviations/Notes:
{any_deviation_notes}
📊 Decision Points:
- Decision 1: {status} - {choice_made}
- Decision 2: Pending
💡 Next Action:
Based on the plan, you should {recommended_action}
```
## Success Criteria
The update is successful when:
1. Plan accurately reflects current workflow state
2. All updates are clearly timestamped
3. Deviations are documented with reasons
4. Progress calculation is correct
5. Next steps are clear to user
6. Plan remains readable and well-formatted
## Error Handling
- **Plan file not found**: Offer to create new plan
- **Malformed plan**: Attempt basic updates, warn user
- **Write permission error**: Show changes that would be made
- **Step not found**: Show available steps, ask for clarification
- **Concurrent updates**: Implement simple locking or warn about conflicts
## Notes
- Always preserve plan history (don't delete old information)
- Keep updates atomic to prevent corruption
- Consider creating backup before major updates
- Updates should enhance, not complicate, the workflow experience
- If plan becomes too cluttered, suggest creating fresh plan for next phase
==================== END: tasks#update-workflow-plan ====================
==================== START: data#bmad-kb ====================
# BMAD Knowledge Base
@@ -737,6 +1314,7 @@ 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.
@@ -922,6 +1500,7 @@ 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
@@ -1391,6 +1970,232 @@ Use the **expansion-creator** pack to build your own:
- **Contributing**: See `CONTRIBUTING.md` for full guidelines
==================== END: data#bmad-kb ====================
==================== START: utils#plan-management ====================
# Plan Management Utility
## Purpose
Provides utilities for agents and tasks to interact with workflow plans, check progress, update status, and ensure workflow steps are executed in the appropriate sequence.
## Core Functions
### 1. Check Plan Existence
[[LLM: When any agent starts or task begins, check if a workflow plan exists]]
```
Check for workflow plan:
1. Look for docs/workflow-plan.md (default location)
2. Check core-config.yml for custom plan location
3. Return plan status (exists/not exists)
```
### 2. Parse Plan Status
[[LLM: Extract current progress from the plan document]]
**Plan Parsing Logic:**
1. **Identify Step Structure**:
- Look for checkbox lines: `- [ ]` or `- [x]`
- Extract step IDs from comments: `<!-- step-id: X.Y -->`
- Identify agent assignments: `<!-- agent: pm -->`
2. **Determine Current State**:
- Last completed step (highest numbered `[x]`)
- Next expected step (first `[ ]` after completed steps)
- Overall progress percentage
3. **Extract Metadata**:
- Workflow type from plan header
- Decision points and their status
- Any deviation notes
### 3. Sequence Validation
[[LLM: Check if requested action aligns with plan sequence]]
**Validation Rules:**
1. **Strict Mode** (enforceSequence: true):
- Must complete steps in exact order
- Warn and block if out of sequence
- Require explicit override justification
2. **Flexible Mode** (enforceSequence: false):
- Warn about sequence deviation
- Allow with confirmation
- Log deviation reason
**Warning Templates:**
```
SEQUENCE WARNING:
The workflow plan shows you should complete "{expected_step}" next.
You're attempting to: "{requested_action}"
In strict mode: Block and require plan update
In flexible mode: Allow with confirmation
```
### 4. Plan Update Operations
[[LLM: Provide consistent way to update plan progress]]
**Update Actions:**
1. **Mark Step Complete**:
- Change `- [ ]` to `- [x]`
- Add completion timestamp comment
- Update any status metadata
2. **Add Deviation Note**:
- Insert note explaining why sequence changed
- Reference the deviation in plan
3. **Update Current Step Pointer**:
- Add/move `<!-- current-step -->` marker
- Update last-modified timestamp
### 5. Integration Instructions
[[LLM: How agents and tasks should use this utility]]
**For Agents (startup sequence)**:
```
1. Check if plan exists using this utility
2. If exists:
- Parse current status
- Show user: "Active workflow plan detected. Current step: {X}"
- Suggest: "Next recommended action: {next_step}"
3. Continue with normal startup
```
**For Tasks (pre-execution)**:
```
1. Check if plan exists
2. If exists:
- Verify this task aligns with plan
- If not aligned:
- In strict mode: Show warning and stop
- In flexible mode: Show warning and ask for confirmation
3. After task completion:
- Update plan if task was a planned step
- Add note if task was unplanned
```
### 6. Plan Status Report Format
[[LLM: Standard format for showing plan status]]
```
📋 Workflow Plan Status
━━━━━━━━━━━━━━━━━━━━
Workflow: {workflow_name}
Progress: {X}% complete ({completed}/{total} steps)
✅ Completed:
- {completed_step_1}
- {completed_step_2}
🔄 Current Step:
- {current_step_description}
📌 Upcoming:
- {next_step_1}
- {next_step_2}
⚠️ Notes:
- {any_deviations_or_notes}
```
### 7. Decision Point Handling
[[LLM: Special handling for workflow decision points]]
When encountering a decision point in the plan:
1. **Identify Decision Marker**: `<!-- decision: {decision_id} -->`
2. **Check Decision Status**: Made/Pending
3. **If Pending**:
- Block progress until decision made
- Show options to user
- Record decision when made
4. **If Made**:
- Verify current path aligns with decision
- Warn if attempting alternate path
### 8. Plan Abandonment
[[LLM: Graceful handling when user wants to stop following plan]]
If user wants to abandon plan:
1. Confirm abandonment intent
2. Add abandonment note to plan
3. Mark plan as "Abandoned" in header
4. Stop plan checking for remainder of session
5. Suggest creating new plan if needed
## Usage Examples
### Example 1: Agent Startup Check
```
BMad Master starting...
[Check for plan]
Found active workflow plan: brownfield-fullstack
Progress: 40% complete (4/10 steps)
Current step: Create PRD (pm agent)
Suggestion: Based on your plan, you should work with the PM agent next.
Use *agent pm to switch, or *plan-status to see full progress.
```
### Example 2: Task Sequence Warning
```
User: *task create-next-story
[Plan check triggered]
⚠️ SEQUENCE WARNING:
Your workflow plan indicates the PRD hasn't been created yet.
Creating stories before the PRD may lead to incomplete requirements.
Would you like to:
1. Continue anyway (will note deviation in plan)
2. Switch to creating PRD first (*agent pm)
3. View plan status (*plan-status)
```
### Example 3: Automatic Plan Update
```
[After completing create-doc task for PRD]
✅ Plan Updated: Marked "Create PRD" as complete
📍 Next step: Create Architecture Document (architect agent)
```
## Implementation Notes
- This utility should be lightweight and fast
- Plan parsing should be resilient to format variations
- Always preserve user agency - warnings not blocks (unless strict mode)
- Plan updates should be atomic to prevent corruption
- Consider plan versioning for rollback capability
## Error Handling
- Missing plan: Return null, don't error
- Malformed plan: Warn but continue, treat as no plan
- Update failures: Log but don't block task completion
- Parse errors: Fallback to basic text search
==================== END: utils#plan-management ====================
==================== START: utils#workflow-management ====================
# Workflow Management
@@ -2901,6 +3706,22 @@ To identify the next logical story based on project progress and epic definition
- `architecture.architectureSharded`: Whether architecture is sharded
- `architecture.architectureFile`: Location of monolithic architecture
- `architecture.architectureShardedLocation`: Location of sharded architecture files
- `workflow.trackProgress`: Whether workflow plan tracking is enabled
- `workflow.planFile`: Location of workflow plan (if tracking enabled)
### 0.5 Check Workflow Plan (if configured)
[[LLM: Check if workflow plan tracking is enabled]]
- If `workflow.trackProgress: true`, check for active plan at `workflow.planFile`
- If plan exists:
- Parse plan to check if story creation is the expected next step
- If out of sequence and `workflow.enforceSequence: true`:
- Show warning: "The workflow plan indicates you should complete {expected_step} before creating stories."
- Block execution unless user explicitly overrides
- If out of sequence and `workflow.enforceSequence: false`:
- Show warning but allow continuation with confirmation
- Continue with story identification after plan check
### 1. Identify Next Story for Preparation
@@ -3124,6 +3945,15 @@ Provide a summary to the user including:
- Recommendations for story review before approval
- Next steps: Story should be reviewed by PO for approval before dev work begins
### 10. Update Workflow Plan (if applicable)
[[LLM: After successful story creation]]
- If `workflow.trackProgress: true` and `workflow.updateOnCompletion: true`:
- Call update-workflow-plan task to mark story creation step complete
- Parameters: task: create-next-story, step_id: {from plan}, status: complete
- If plan shows next step, mention it in completion message
[[LLM: Remember - The success of this task depends on extracting real, specific technical details from the architecture shards. The dev agent should have everything they need in the story file without having to search through multiple documents.]]
==================== END: tasks#create-next-story ====================

1183
dist/teams/team-no-ui.txt vendored
View File

File diff suppressed because it is too large Load Diff

14
tools/installer/lib/installer.js
View File

@@ -757,12 +757,7 @@ class Installer {
const ides = manifest.ides_setup || [];
if (ides.includes('cursor')) {
console.log(chalk.yellow.bold("\n⚠️ IMPORTANT: Cursor Custom Modes Update Required"));
console.log(chalk.yellow("Since agent files have been repaired, you need to manually update your Cursor custom modes:"));
console.log(chalk.yellow("1. Open Cursor Settings (Cmd/Ctrl + ,)"));
console.log(chalk.yellow("2. Go to: Features > Cursor Tab > Custom Modes"));
console.log(chalk.yellow("3. Update each custom mode with the latest agent templates from:"));
console.log(chalk.yellow(` ${path.join(installDir, '.bmad-core', 'agents')}`));
console.log(chalk.yellow("4. Copy the full content of each agent file into the corresponding custom mode"));
console.log(chalk.yellow("Since agent files have been repaired, you need to update any custom agent modes configured in the Cursor custom agent GUI per the Cursor docs."));
}
} catch (error) {
@@ -861,12 +856,7 @@ class Installer {
// Warning for Cursor custom modes if agents were updated
if (options.isUpdate && ides.includes('cursor')) {
console.log(chalk.yellow.bold("\n⚠️ IMPORTANT: Cursor Custom Modes Update Required"));
console.log(chalk.yellow("Since agents have been updated, you need to manually update your Cursor custom modes:"));
console.log(chalk.yellow("1. Open Cursor Settings (Cmd/Ctrl + ,)"));
console.log(chalk.yellow("2. Go to: Features > Cursor Tab > Custom Modes"));
console.log(chalk.yellow("3. Update each custom mode with the latest agent templates from:"));
console.log(chalk.yellow(` ${path.join(installDir, '.bmad-core', 'agents')}`));
console.log(chalk.yellow("4. Copy the full content of each agent file into the corresponding custom mode"));
console.log(chalk.yellow("Since agents have been updated, you need to update any custom agent modes configured in the Cursor custom agent GUI per the Cursor docs."));
}
}