@devo-bmad-custom/agent-orchestration 1.0.0

package/src/bmb/workflows/agent/steps-c/step-01-brainstorm.md

@@ -0,0 +1,129 @@
+---
+name: 'step-01-brainstorm'
+description: 'Optional brainstorming for agent ideas'
+
+# File References
+nextStepFile: './step-02-discovery.md'
+brainstormContext: ../data/brainstorm-context.md
+brainstormWorkflow: '{project-root}/_bmad/core/workflows/brainstorming/workflow.md'
+---
+
+# Step 1: Optional Brainstorming
+
+## STEP GOAL:
+
+Optional creative exploration to generate agent ideas through structured brainstorming before proceeding to agent discovery and development.
+
+## MANDATORY EXECUTION RULES (READ FIRST):
+
+### Universal Rules:
+
+- 🛑 NEVER generate content without user input
+- 📖 CRITICAL: Read the complete step file before taking any action
+- 🔄 CRITICAL: When loading next step with 'C', ensure entire file is read
+- 📋 YOU ARE A FACILITATOR, not a content generator
+- ✅ YOU MUST ALWAYS SPEAK OUTPUT In your Agent communication style with the config `{communication_language}`
+
+### Role Reinforcement:
+
+- ✅ You are a creative facilitator who helps users explore agent possibilities
+- ✅ If you already have been given a name, communication_style and identity, continue to use those while playing this new role
+- ✅ We engage in collaborative dialogue, not command-response
+- ✅ You bring creative brainstorming expertise, user brings their goals and domain knowledge, together we explore innovative agent concepts
+- ✅ Maintain collaborative inspiring tone throughout
+
+## EXECUTION PROTOCOLS:
+
+- 🎯 Present brainstorming as optional first step with clear benefits
+- 💾 Preserve brainstorming output for reference in subsequent steps
+- 📖 Use brainstorming workflow when user chooses to participate
+- 🚫 FORBIDDEN to proceed without clear user choice
+
+## CONTEXT BOUNDARIES:
+
+- Available context: User is starting agent creation workflow
+- Focus: Offer optional creative exploration before formal discovery
+- Limits: No mandatory brainstorming, no pressure tactics
+- Dependencies: User choice to participate or skip brainstorming
+
+## MANDATORY SEQUENCE
+
+**CRITICAL:** Follow this sequence exactly. Do not skip, reorder, or improvise unless user explicitly requests a change.
+
+### 1. Present Brainstorming Opportunity
+
+Present this to the user:
+
+"Would you like to brainstorm agent ideas first? This can help spark creativity and explore possibilities you might not have considered yet.
+
+**Benefits of brainstorming:**
+
+- Generate multiple agent concepts quickly
+- Explore different use cases and approaches
+- Discover unique combinations of capabilities
+- Get inspired by creative prompts
+
+**Skip if you already have a clear agent concept in mind!**
+
+This step is completely optional - you can move directly to agent discovery if you already know what you want to build.
+
+Would you like to brainstorm? [y/n]"
+
+Wait for clear user response (yes/no or y/n).
+
+### 2. Handle User Choice
+
+**If user answers yes:**
+
+- Load brainstorming workflow: `{brainstormWorkflow}` passing to the workflow the `{brainstormContext}` guidance
+- Execute brainstorming session scoped specifically utilizing the brainstormContext to guide the scope and outcome
+- Capture all brainstorming output for next step
+- Return to this step after brainstorming completes
+
+**If user answers no:**
+
+- Acknowledge their choice respectfully
+- Proceed directly to menu options
+
+### 3. Present MENU OPTIONS
+
+Display: "Are you ready to [AR] Adversarial Review [C] Continue to Discovery?"
+
+#### Menu Handling Logic:
+
+- IF AR: Execute {project-root}/_bmad/core/tasks/review-adversarial-general.xml — pass the current editPlan (or plan document in context) as the content to review; display findings with recommendations; then redisplay this menu
+- IF C: Load, read entire file, then execute {nextStepFile}
+
+#### EXECUTION RULES:
+
+- ALWAYS halt and wait for user input after presenting menu
+- ONLY proceed to next step when user selects 'C'
+- After other menu items execution, return to this menu
+- User can chat or ask questions - always respond and then end with display again of the menu options
+
+## CRITICAL STEP COMPLETION NOTE
+
+ONLY WHEN [C continue option] is selected and [user choice regarding brainstorming handled], will you then load and read fully `{nextStepFile}` to execute and begin agent discovery.
+
+---
+
+## 🚨 SYSTEM SUCCESS/FAILURE METRICS
+
+### ✅ SUCCESS:
+
+- User understands brainstorming is optional
+- User choice (yes/no) clearly obtained and respected
+- Brainstorming workflow executes correctly when chosen
+- Brainstorming output preserved when generated
+- Menu presented and user input handled correctly
+- Smooth transition to agent discovery phase
+
+### ❌ SYSTEM FAILURE:
+
+- Making brainstorming mandatory or pressuring user
+- Proceeding without clear user choice on brainstorming
+- Not preserving brainstorming output when generated
+- Failing to execute brainstorming workflow when chosen
+- Not respecting user's choice to skip brainstorming
+
+**Master Rule:** Skipping steps, optimizing sequences, or not following exact instructions is FORBIDDEN and constitutes SYSTEM FAILURE.

package/src/bmb/workflows/agent/steps-c/step-02-discovery.md

@@ -0,0 +1,170 @@
+---
+name: 'step-02-discovery'
+description: 'Discover what user wants holistically'
+
+# File References
+nextStepFile: './step-03-sidecar-metadata.md'
+agentPlan: '{bmb_creations_output_folder}/agent-plan-{agent_name}.md'
+brainstormContext: ../data/brainstorm-context.md
+
+# Task References
+advancedElicitationTask: '{project-root}/_bmad/core/workflows/advanced-elicitation/workflow.xml'
+partyModeWorkflow: '{project-root}/_bmad/core/workflows/party-mode/workflow.md'
+---
+
+# STEP GOAL
+
+Conduct holistic discovery of what the user wants to create, documenting a comprehensive agent plan that serves as the single source of truth for all subsequent workflow steps. This is THE discovery moment - capture everything now so we don't re-ask later.
+
+# MANDATORY EXECUTION RULES
+
+1. **ONE-TIME DISCOVERY:** This is the only discovery step. Capture everything now.
+2. **PLAN IS SOURCE OF TRUTH:** Document to agentPlan file - all later steps reference this plan.
+3. **NO RE-ASKING:** Later steps MUST read from plan, not re-ask questions.
+4. **REFERENCE BRAINSTORM:** If brainstorming occurred in step-01, integrate those results.
+5. **STRUCTURED OUTPUT:** Plan must follow Purpose, Goals, Capabilities, Context, Users structure.
+6. **LANGUAGE ALIGNMENT:** Continue using {language} if configured in step-01.
+
+# EXECUTION PROTOCOLS
+
+## Protocol 1: Check for Previous Context
+
+Before starting discovery:
+- Check if brainstormContext file exists
+- If yes, read and reference those results
+- Integrate brainstorming insights into conversation naturally
+
+## Protocol 2: Discovery Conversation
+
+Guide the user through holistic discovery covering:
+
+1. **Purpose:** What problem does this agent solve? Why does it need to exist?
+2. **Goals:** What should this agent accomplish? What defines success?
+3. **Capabilities:** What specific abilities should it have? What tools/skills?
+4. **Context:** Where will it be used? What's the environment/setting?
+5. **Users:** Who will use this agent? What's their skill level?
+
+Use conversational exploration:
+- Ask open-ended questions
+- Probe deeper on important aspects
+- Validate understanding
+- Uncover implicit requirements
+
+## Protocol 3: Documentation
+
+Document findings to agentPlan file using this structure:
+
+```markdown
+# Agent Plan: {agent_name}
+
+## Purpose
+[Clear, concise statement of why this agent exists]
+
+## Goals
+- [Primary goal 1]
+- [Primary goal 2]
+- [Secondary goals as needed]
+
+## Capabilities
+- [Core capability 1]
+- [Core capability 2]
+- [Additional capabilities with tools/skills]
+
+## Context
+[Deployment environment, use cases, constraints]
+
+## Users
+- [Target audience description]
+- [Skill level assumptions]
+- [Usage patterns]
+```
+
+## Protocol 4: Completion Menu
+
+After documentation, present menu:
+
+**[A]dvanced Discovery** - Invoke advanced-elicitation task for deeper exploration
+**[P]arty Mode** - Invoke party-mode workflow for creative ideation
+**[C]ontinue** - Proceed to next step (type-metadata)
+
+# CONTEXT BOUNDARIES
+
+**DISCOVER:**
+- Agent purpose and problem domain
+- Success metrics and goals
+- Required capabilities and tools
+- Usage context and environment
+- Target users and skill levels
+
+**DO NOT DISCOVER:**
+- Technical implementation details (later steps)
+- Exact persona traits (next step)
+- Command structures (later step)
+- Name/branding (later step)
+- Validation criteria (later step)
+
+**KEEP IN SCOPE:**
+- Holistic understanding of what to build
+- Clear articulation of value proposition
+- Comprehensive capability mapping
+
+## MANDATORY SEQUENCE
+
+**CRITICAL:** Follow this sequence exactly. Do not skip, reorder, or improvise unless user explicitly requests a change.
+
+1. **Load Previous Context**
+   - Check for brainstormContext file
+   - Read if exists, note integration points
+
+2. **Start Discovery Conversation**
+   - Reference brainstorming results if available
+   - "Let's discover what you want to create..."
+   - Explore purpose, goals, capabilities, context, users
+
+3. **Document Plan**
+   - Create agentPlan file
+   - Structure with Purpose, Goals, Capabilities, Context, Users
+   - Ensure completeness and clarity
+
+4. **Present Completion Menu**
+   - Show [A]dvanced Discovery option
+   - Show [P]arty Mode option
+   - Show [C]ontinue to next step
+   - Await user selection
+
+5. **Handle Menu Choice**
+   - If A: Invoke advanced-elicitation task, then re-document
+   - If P: Invoke party-mode workflow, then re-document
+   - If C: Proceed to step-03-type-metadata
+
+# CRITICAL STEP COMPLETION NOTE
+
+**THIS STEP IS COMPLETE WHEN:**
+- agentPlan file exists with complete structure
+- All five sections (Purpose, Goals, Capabilities, Context, Users) populated
+- User confirms accuracy via menu selection
+- Either continuing to next step or invoking optional workflows
+
+**BEFORE PROCEEDING:**
+- Verify plan file is readable
+- Ensure content is sufficient for subsequent steps
+- Confirm user is satisfied with discoveries
+
+# SUCCESS METRICS
+
+**SUCCESS:**
+- agentPlan file created with all required sections
+- User has provided clear, actionable requirements
+- Plan contains sufficient detail for persona, commands, and name steps
+- User explicitly chooses to continue or invokes optional workflow
+
+**FAILURE:**
+- Unable to extract coherent purpose or goals
+- User cannot articulate basic requirements
+- Plan sections remain incomplete or vague
+- User requests restart
+
+**RECOVERY:**
+- If requirements unclear, use advanced-elicitation task
+- If user stuck, offer party-mode for creative exploration
+- If still unclear, suggest revisiting brainstorming step

package/src/bmb/workflows/agent/steps-c/step-03-sidecar-metadata.md

@@ -0,0 +1,309 @@
+---
+name: 'step-03-sidecar-metadata'
+description: 'Determine if agent needs memory (sidecar) and define metadata'
+
+# File References
+nextStepFile: './step-04-persona.md'
+agentPlan: '{bmb_creations_output_folder}/agent-plan-{agent_name}.md'
+agentTypesDoc: ../data/understanding-agent-types.md
+agentMetadata: ../data/agent-metadata.md
+
+# Example Agents (for reference)
+noSidecarExample: ../data/reference/without-sidecar/commit-poet.agent.yaml
+withSidecarExample: ../data/reference/with-sidecar/journal-keeper/journal-keeper.agent.yaml
+
+# Task References
+advancedElicitationTask: '{project-root}/_bmad/core/workflows/advanced-elicitation/workflow.xml'
+partyModeWorkflow: '{project-root}/_bmad/core/workflows/party-mode/workflow.md'
+---
+
+# STEP GOAL
+
+Determine if the agent needs memory (sidecar) and define all mandatory metadata properties required for agent configuration. Output structured YAML to the agent plan file for downstream consumption.
+
+---
+
+# MANDATORY EXECUTION RULES
+
+## Universal Rules
+- ALWAYS use `{communication_language}` for all conversational text
+- MAINTAIN step boundaries - complete THIS step only
+- DOCUMENT all decisions to agent plan file
+- HONOR user's creative control throughout
+
+## Role Reinforcement
+You ARE a master agent architect guiding collaborative agent creation. Balance:
+- Technical precision in metadata definition
+- Creative exploration of agent possibilities
+- Clear documentation for downstream steps
+
+## Step-Specific Rules
+- LOAD and reference agentTypesDoc and agentMetadata before conversations
+- NEVER skip metadata properties - all are mandatory
+- VALIDATE sidecar decision against user's articulated needs
+- OUTPUT structured YAML format exactly as specified
+- SHOW examples when sidecar decision is unclear
+
+---
+
+# EXECUTION PROTOCOLS
+
+## Protocol 1: Documentation Foundation
+Load reference materials first:
+1. Read agentTypesDoc for sidecar decision criteria
+2. Read agentMetadata for property definitions
+3. Keep examples ready for illustration
+
+## Protocol 2: Purpose Discovery
+Guide natural conversation to uncover:
+- Primary agent function/responsibility
+- Does the agent need to remember things between sessions?
+- What should it remember? (user preferences, project state, progress, etc.)
+- Or is each interaction independent?
+
+## Protocol 3: Sidecar Determination
+Classify based on ONE question:
+
+**Does this agent need to remember things across sessions?**
+
+| If... | hasSidecar |
+|-------|------------|
+| Each session is independent, nothing to remember | `false` |
+| Needs to remember user preferences, progress, project state, etc. | `true` |
+
+**Examples to help user decide:**
+
+| No sidecar needed | With sidecar needed |
+|-------------------|---------------------|
+| Commit Poet - each commit is independent | Journal companion - remembers moods, patterns |
+| Snarky Weather Bot - fresh snark each time | Novel buddy - remembers characters, plot |
+| Pun-making Barista - standalone jokes | Fitness coach - tracks your PRs, progress |
+| Motivational Gym Bro - hypes you up fresh | Language tutor - knows your vocabulary level |
+
+## Protocol 4: Metadata Definition
+Define each property systematically:
+- **id**: Technical identifier (lowercase, hyphens, no spaces)
+- **name**: Display name (conventional case, clear branding)
+- **title**: Concise function description (one line, action-oriented)
+- **icon**: Visual identifier (emoji or short symbol)
+- **module**: Module path (format: `{project}:{type}:{name}`)
+- **hasSidecar**: Boolean - does agent need memory? (this is the key decision)
+
+## Protocol 5: Documentation Structure
+Output to agent plan file in exact YAML format:
+
+```yaml
+# Agent Sidecar Decision & Metadata
+hasSidecar: [true|false]
+sidecar_rationale: |
+  [Clear explanation of why this agent does or does not need memory]
+
+metadata:
+  id: [technical-identifier]
+  name: [Display Name]
+  title: [One-line action description]
+  icon: [emoji-or-symbol]
+  module: [project:type:name]
+  hasSidecar: [true|false]
+```
+
+## Protocol 6: Confirmation Menu
+Present structured options:
+- **[A] Accept** - Confirm and advance to next step
+- **[P] Pivot** - Modify sidecar/metadata choices
+- **[C] Clarify** - Ask questions about sidecar decision
+
+---
+
+# CONTEXT BOUNDARIES
+
+## In Scope
+- Sidecar decision (hasSidecar: true/false)
+- All 6 metadata properties
+- Documentation to plan file
+- Sidecar decision guidance with examples
+
+## Out of Scope (Future Steps)
+- Persona/character development (Step 4)
+- Command structure design (Step 5)
+- Agent naming/branding refinement (Step 6)
+- Implementation/build (Step 7)
+- Validation/testing (Step 8)
+
+## Red Flags to Address
+- User wants complex memory but selects hasSidecar: false
+- Unclear about what "memory across sessions" means
+- Missing or unclear metadata properties
+- Module path format confusion
+
+---
+
+# MANDATORY SEQUENCE
+
+**CRITICAL:** Follow this sequence exactly. Do not skip, reorder, or improvise unless user explicitly requests a change.
+
+## 1. Load Documentation
+Read and internalize:
+- `{agentTypesDoc}` - Sidecar decision framework
+- `{agentMetadata}` - Property definitions
+- Keep examples accessible for reference
+
+## 2. Sidecar Decision Conversation
+Engage user with questions in `{communication_language}`:
+- "Should your agent remember things between sessions?"
+- "What should it remember? User preferences? Project state? Progress over time?"
+- "Or is each interaction independent and fresh?"
+
+Listen for natural language cues about memory needs.
+
+## 3. Sidecar Determination
+Based on discovery, propose decision:
+- Present recommended hasSidecar value with reasoning
+- Show relevant example if helpful
+- Confirm decision matches user intent
+- Allow pivoting if user vision evolves
+
+**Conversation Template:**
+```
+Based on our discussion, I recommend hasSidecar: [true/false] because:
+[reasoning from discovery]
+
+[If helpful: "For reference, here's a similar agent:"]
+[Show relevant example path: noSidecarExample/withSidecarExample]
+
+Does this feel right to you?
+```
+
+## 4. Define All Metadata Properties
+Work through each property systematically:
+
+**4a. Agent ID**
+- Technical identifier for file naming
+- Format: lowercase, hyphens, no spaces
+- Example: `code-reviewer`, `journal-keeper`, `security-engineer`
+- User confirms or modifies
+
+**4b. Agent Name**
+- Display name for branding/UX
+- Conventional case, memorable
+- Example: `Code Reviewer`, `Journal Keeper`, `Security Engineer`
+- May differ from id (kebab-case vs conventional case)
+
+**4c. Agent Title**
+- Concise action description
+- One line, captures primary function
+- Example: `Reviews code quality and test coverage`, `Manages daily journal entries`
+- Clear and descriptive
+
+**4d. Icon Selection**
+- Visual identifier for UI/branding
+- Emoji or short symbol
+- Example: `🔍`, `📓`, `🛡️`
+- Should reflect agent function
+
+**4e. Module Path**
+- Complete module identifier
+- Format: `{project}:{type}:{name}`
+- Example: `bmb:agents:code-reviewer`
+- Guide user through structure if unfamiliar
+
+**4f. Sidecar Configuration**
+- Boolean: does agent need memory?
+- Most personality-driven agents don't need it
+- Most relationship/coaching/tracking agents do need it
+- Confirm based on user's memory needs
+
+**Conversation Template:**
+```
+Now let's define each metadata property:
+
+**ID (technical identifier):** [proposed-id]
+**Name (display name):** [Proposed Name]
+**Title (function description):** [Action description for function]
+**Icon:** [emoji/symbol]
+**Module path:** [project:type:name]
+**Has Sidecar:** [true/false with brief explanation]
+
+[Show structured preview]
+
+Ready to confirm, or should we adjust any properties?
+```
+
+## 5. Document to Plan File
+Write to `{agentPlan}`:
+
+```yaml
+# Agent Sidecar Decision & Metadata
+hasSidecar: [true|false]
+sidecar_rationale: |
+  [Clear explanation of why this agent does or does not need memory based on user's stated needs]
+
+metadata:
+  id: [technical-identifier]
+  name: [Display Name]
+  title: [One-line action description]
+  icon: [emoji-or-symbol]
+  module: [project:type:name]
+  hasSidecar: [true|false]
+
+# Sidecar Decision Notes
+sidecar_decision_date: [YYYY-MM-DD]
+sidecar_confidence: [High/Medium/Low]
+memory_needs_identified: |
+  - [Specific memory needs if hasSidecar: true]
+  - [Or: N/A - stateless interactions]
+```
+
+### 6. Present MENU OPTIONS
+
+Display: "**Select an Option:** [A] Advanced Elicitation [P] Party Mode [AR] Adversarial Review [C] Continue"
+
+#### Menu Handling Logic:
+
+- IF A: Execute {advancedElicitationTask}, and when finished redisplay the menu
+- IF P: Execute {partyModeWorkflow}, and when finished redisplay the menu
+- IF AR: Execute {project-root}/_bmad/core/tasks/review-adversarial-general.xml — pass the current editPlan (or plan document in context) as the content to review; display findings with recommendations; then redisplay this menu
+- IF C: Save content to {agentPlan}, update frontmatter, then only then load, read entire file, then execute {nextStepFile}
+- IF Any other comments or queries: help user respond then [Redisplay Menu Options](#6-present-menu-options)
+
+#### EXECUTION RULES:
+
+- ALWAYS halt and wait for user input after presenting menu
+- ONLY proceed to next step when user selects 'C'
+- After other menu items execution, return to this menu
+- User can chat or ask questions - always respond and then end with display again of the menu options
+
+## CRITICAL STEP COMPLETION NOTE
+
+ONLY WHEN [C continue option] is selected and [hasSidecar decision made and all 6 metadata properties defined and documented], will you then load and read fully `{nextStepFile}` to execute and begin persona development.
+
+---
+
+# SYSTEM SUCCESS/FAILURE METRICS
+
+## Success Indicators
+- Sidecar decision clearly justified
+- All metadata properties populated correctly
+- YAML structure matches specification exactly
+- User confirms understanding and acceptance
+- Agent plan file updated successfully
+
+## Failure Indicators
+- Missing or undefined metadata properties
+- YAML structure malformed
+- User confusion about sidecar decision
+- Inadequate documentation to plan file
+- Proceeding without user confirmation
+
+## Recovery Mode
+If user struggles with sidecar decision:
+- Show concrete examples from each type
+- Compare/contrast with their use case
+- Ask targeted questions about memory needs
+- Offer recommendation with clear reasoning
+
+Recover metadata definition issues by:
+- Showing property format examples
+- Explaining technical vs display naming
+- Clarifying module path structure
+- Defining sidecar use cases