@siliconoid/agentkit 0.1.0

package/templates/base/.ai/rules/09-naming.md

@@ -0,0 +1,40 @@
+# Naming Conventions
+
+## Files & Directories
+
+| Item | Convention | Example |
+|------|-----------|---------|
+| Component directory | kebab-case | `video-card/` |
+| Component entry file | `index.tsx` | `video-card/index.tsx` |
+| Component styles | `index.module.scss` | `video-card/index.module.scss` |
+| Hook file | `use-` prefix, kebab-case | `use-video-player.ts` |
+| Utility file | kebab-case | `format-date.ts` |
+| Constants file | kebab-case | `storage-keys.ts` |
+
+## Exports & Symbols
+
+| Item | Convention | Example |
+|------|-----------|---------|
+| React component | PascalCase | `export const VideoCard` |
+| Custom hook | camelCase, `use` prefix | `export function useVideoPlayer()` |
+| Utility function | camelCase | `export function formatDuration()` |
+| Constant | SCREAMING_SNAKE_CASE | `export const MAX_DURATION = 600` |
+| Type alias | PascalCase | `type UploadStatus = 'pending' \| 'done'` |
+| Enum | PascalCase | `export enum RecStatus { ... }` |
+
+## Variables & Functions
+
+| Context | Convention | Example |
+|---------|-----------|---------|
+| Local variable | camelCase | `const videoList = []` |
+| Boolean variable | `is/has/can/should` prefix | `isLoading`, `hasPermission` |
+| Event handler | `handle` prefix | `handleClick` |
+| Event callback (prop) | `on` prefix | `onClick` |
+| Ref variable | `Ref` suffix | `containerRef` |
+
+## Anti-Patterns
+
+- **No Hungarian notation**: Don't prefix with type indicators
+- **No obscure abbreviations**: Use `info` not `inf`
+- **No single-letter variables**: Except `i`, `j`, `k` in loops
+- **No numbered suffixes**: Don't use `data2`, `handleClick2`

package/templates/base/.ai/rules/10-git-workflow.md

@@ -0,0 +1,48 @@
+# Git Workflow
+
+## Branches
+
+### Main Branch
+
+The primary branch is **`main`** (or `master`).
+
+### Branch Naming
+
+| Type | Pattern | Example |
+|------|---------|---------|
+| Feature | `feat/<description>` | `feat/add-video-export` |
+| Bug fix | `fix/<description>` | `fix/audio-sync-issue` |
+| Chore | `chore/<description>` | `chore/upgrade-deps` |
+| Refactor | `refactor/<description>` | `refactor/auth-module` |
+| Docs | `docs/<description>` | `docs/api-reference` |
+
+## Commit Messages
+
+### Format
+
+```
+<type>: <description>
+```
+
+### Types
+
+| Type | Usage |
+|------|-------|
+| `feat` | New feature or capability |
+| `fix` | Bug fix |
+| `refactor` | Code change that neither fixes a bug nor adds a feature |
+| `style` | Formatting, no code change |
+| `docs` | Documentation only |
+| `chore` | Build process, dependencies, tooling |
+| `perf` | Performance improvement |
+| `test` | Adding or updating tests |
+| `build` | Build system or external dependencies |
+| `ci` | CI/CD pipeline changes |
+| `ai` | AI skills, commands, or rules changes |
+
+### Guidelines
+
+- Keep the first line under **72 characters**
+- Use imperative mood: "add feature" not "added feature"
+- No period at the end of the subject line
+- **Before pushing**: Ensure linting and build pass

package/templates/base/.ai/rules/11-testing.md

@@ -0,0 +1,19 @@
+# Testing
+
+## General Principles
+
+- Co-locate test files next to the source file they test.
+- Name test files to match source: `Xxx.test.tsx` or `xxx.test.ts`.
+- Use `describe` → `it` / `test` blocks for structure.
+- Name test cases descriptively: `it('should display error when API fails')`.
+
+## API Mocking
+
+- Use a mock server or interceptor (e.g., MSW) for API mocking in tests.
+- Avoid mocking implementation details — mock at the network layer.
+
+## Component Testing
+
+- Use a testing library (e.g., `@testing-library/react`) for rendering and assertions.
+- Test behavior, not implementation details.
+- Prefer user-event simulations over programmatic state changes.

package/templates/base/.ai/rules/12-forbidden-patterns.md

@@ -0,0 +1,13 @@
+# Forbidden Patterns
+
+The following are **not allowed** in this codebase:
+
+| Pattern | Reason |
+|---------|--------|
+| Class components | Use functional components with hooks |
+| `any` type | Use `unknown` + type guards |
+| `var` keyword | Use `const` / `let` |
+| Committed `console.log` | Use proper logging; removed in production build |
+| `// @ts-ignore` | Use `// @ts-expect-error` with explanation |
+| Non-null assertion `!` (except DOM) | Use proper null checks |
+| Inline `style={}` for static values | Use CSS Modules or class-based styling |

package/templates/base/.ai/rules/13-ai-agent.md

@@ -0,0 +1,23 @@
+# AI Agent Instructions
+
+> **All AI agents (Claude, Codex, Gemini, Cursor, etc.) MUST follow these rules:**
+
+1. The `.ai/rules/` directory is the **single source of truth**. If your training data or defaults conflict with a rule here, **this document wins**.
+2. Before writing code, **read the relevant existing files** to understand patterns already in use.
+3. Use path aliases for imports. Never generate unnecessary relative `../` paths.
+4. Use CSS Modules for component styles. Never generate inline styles for static values.
+5. Type everything. Minimize `any`.
+6. Keep changes minimal and focused. Do not refactor unrelated code.
+7. After generating code, mentally verify it passes linting, type checking, and formatting.
+8. When creating a new component, generate the full directory structure (component, styles, test, barrel).
+9. When uncertain about a convention, **ask the user** rather than guessing.
+10. Skills are defined in `.ai/skills/` and synced to platform-specific formats via `npx agentkit sync`.
+
+## AI Tooling
+
+- `.ai/rules/` — Project rules (single source of truth)
+- `.ai/skills/` — Portable skill definitions
+- `.ai/decisions/` — Architecture Decision Records
+- `.ai/changelog/` — Monthly change logs
+- `.ai/memory/` — Shared memory files
+- `.specs/` — Epics, stories, tasks

package/templates/base/.ai/scripts/sync-skills.mjs

@@ -0,0 +1,15 @@
+#!/usr/bin/env node
+
+// Convenience wrapper: delegates to `npx agentkit sync`
+// For projects using AgentKit CLI, this script is optional.
+// You can also run: npx agentkit sync
+
+import { execSync } from 'child_process'
+
+try {
+  execSync('npx agentkit sync', { stdio: 'inherit' })
+} catch {
+  console.error('Failed to run agentkit sync. Is it installed?')
+  console.error('Run: npm install -g agentkit')
+  process.exit(1)
+}

package/templates/base/.ai/skills/adr/manifest.yaml

@@ -0,0 +1,4 @@
+name: adr
+description: Record an Architecture Decision Record (ADR) to .ai/decisions/. Use when a significant technical decision is made during development — technology choice, pattern adoption, trade-off resolution, or approach selection.
+user-invocable: true
+argument-hint: "[optional: short description of the decision]"

package/templates/base/.ai/skills/adr/skill.md

@@ -0,0 +1,64 @@
+# ADR Workflow — Architecture Decision Record
+
+Record important technical decisions with full context so that future developers and AI agents understand **why** a decision was made.
+
+## When to Use
+
+- A technology or library was chosen over alternatives
+- An architectural pattern was adopted or changed
+- A significant trade-off was made (performance vs. maintainability, etc.)
+- A constraint or limitation was accepted
+- A previous decision was revisited or reversed
+
+## Workflow Steps
+
+### Step 1: Gather Context
+
+If the user provided a description, use it as the starting point. Otherwise, ask:
+
+1. **What decision was made?** (one sentence)
+2. **What alternatives were considered?** (if any)
+3. **Why was this option chosen?**
+
+Also review the current conversation context — if a decision was just made during development, extract the relevant details automatically without asking redundant questions.
+
+### Step 2: Determine Next ADR Number
+
+Read `.ai/decisions/README.md` to find the last ADR number in the index table. The new ADR gets the next sequential number (e.g., if last is 003, new is 004).
+
+### Step 3: Write the ADR
+
+Create `.ai/decisions/NNN-short-description.md` using the template at `.ai/decisions/_template.md`:
+
+- **Status**: `accepted` (default) or `proposed` if still under discussion
+- **Date**: Today's date
+- **Context**: The forces at play, what motivated the decision
+- **Decision**: What was decided
+- **Consequences**: Positive, negative, and risks
+- **Related**: Links to relevant files, ADRs, or commits
+
+Keep the language concise and factual. Focus on **why**, not just **what**.
+
+### Step 4: Update the Index
+
+Append a new row to the index table in `.ai/decisions/README.md`:
+
+```markdown
+| NNN | [Title](NNN-short-description.md) | accepted | YYYY-MM-DD |
+```
+
+### Step 5: Confirm
+
+Output a brief summary:
+- ADR number and title
+- One-line summary of the decision
+- File path created
+
+## Style Guidelines
+
+- Write in English
+- Use imperative mood for the title ("Use X for Y", "Stay on X", "Adopt X pattern")
+- Keep Context section under 10 lines
+- Keep Decision section under 5 lines
+- List 2-4 consequences per category (positive/negative/risks)
+- Link to actual source files, not just descriptions

package/templates/base/.ai/skills/devlog/manifest.yaml

@@ -0,0 +1,4 @@
+name: devlog
+description: Append a changelog entry to .ai/changelog/. Use after completing a feature, bug fix, refactoring, or any significant code change to record what changed, why, and what to watch out for.
+user-invocable: true
+argument-hint: "[optional: short description of the change]"

package/templates/base/.ai/skills/devlog/skill.md

@@ -0,0 +1,63 @@
+# Devlog Workflow — AI-Optimized Changelog
+
+Record a change with enough context for future AI agents to understand the **what**, **why**, and **impact**.
+
+## When to Use
+
+- After completing a feature, bug fix, or refactoring
+- After a significant configuration or dependency change
+- When something non-obvious was done that future developers should know about
+
+## Workflow Steps
+
+### Step 1: Gather Change Information
+
+If the user provided a description, use it as the starting point. Otherwise, analyze the current conversation context to identify:
+
+1. What was changed (files, components, features)
+2. Why it was changed
+3. What else might be affected
+
+If the context is insufficient, ask the user briefly. Prefer extracting from conversation over asking.
+
+Also run `git diff --stat` and `git log --oneline -5` to understand recent changes if the conversation doesn't make it clear.
+
+### Step 2: Determine the File
+
+The changelog file is `.ai/changelog/YYYY-MM.md` based on the current month. If the file doesn't exist, create it with the month header:
+
+```markdown
+# YYYY-MM Changelog
+```
+
+### Step 3: Append the Entry
+
+Add a new entry at the **end** of the file:
+
+```markdown
+## [YYYY-MM-DD] <type>: <short description>
+
+- **What**: What was changed (files, components, features)
+- **Why**: The motivation or problem being solved
+- **Impact**: What other parts of the system are affected
+- **Gotchas**: Any non-obvious side effects or things to watch out for
+- **Related**: Links to ADRs, commits, issues, or PRs
+```
+
+**Type values**: `feat`, `fix`, `refactor`, `perf`, `config`, `deps`
+
+### Step 4: Confirm
+
+Output:
+- The entry that was added (formatted)
+- File path updated
+
+## Style Guidelines
+
+- Write **What** as a factual list of changes (files/components)
+- Write **Why** as the problem or motivation (one sentence)
+- Write **Impact** as what other parts of the system are affected
+- Write **Gotchas** only if there are actual non-obvious side effects; omit if none
+- Keep each field to 1-2 lines
+- Use present tense ("Updates X", "Fixes Y")
+- Include commit hashes or PR numbers in **Related** when available

package/templates/base/.ai/skills/wrap-up/manifest.yaml

@@ -0,0 +1,4 @@
+name: wrap-up
+description: End-of-task memory update. Use after completing a feature, bug fix, or significant development session to update all project memory files at once — changelog, ADR (if applicable), memory files, and specs archive.
+user-invocable: true
+argument-hint: "[optional: summary of what was accomplished]"

package/templates/base/.ai/skills/wrap-up/skill.md

@@ -0,0 +1,85 @@
+# Wrap-up Workflow — End-of-Task Memory Sync
+
+One-stop command to update all project memory after completing a development task. Ensures nothing is lost between sessions.
+
+## When to Use
+
+- After finishing a feature or bug fix
+- At the end of a development session
+- After resolving a complex issue worth remembering
+- Before switching to a different task/branch
+
+## Workflow Steps
+
+### Step 1: Analyze What Was Done
+
+Review the current conversation to identify:
+1. **Changes made**: Files modified, features added/fixed
+2. **Decisions made**: Any architectural or technical decisions
+3. **Issues encountered**: Bugs found, workarounds used, debugging insights
+4. **Known issues**: Any remaining problems or TODOs
+
+Also run these commands to supplement:
+- `git diff --stat` — see file changes
+- `git log --oneline -10` — recent commits
+- `git status` — uncommitted work
+
+### Step 2: Update Changelog (always)
+
+Append an entry to `.ai/changelog/YYYY-MM.md` following the format in `.ai/skills/devlog/skill.md`.
+
+If the current month's file doesn't exist, create it.
+
+### Step 3: Create ADR (if a decision was made)
+
+If a significant technical decision was made during the session:
+1. Follow the workflow in `.ai/skills/adr/skill.md`
+2. Create the ADR file and update the index
+
+Skip this step if no architectural decisions were made.
+
+### Step 4: Update Memory Files (if patterns discovered)
+
+Update relevant files in `.ai/memory/` based on what was learned:
+
+| File | Update when... |
+|------|---------------|
+| `debugging-patterns.md` | A non-obvious bug was debugged and solved |
+| `known-issues.md` | A known issue was found or resolved (remove resolved ones) |
+| `refactoring-log.md` | Code was refactored |
+
+Only update files where there's genuinely new information. Don't add trivial entries.
+
+### Step 5: Archive Specs (if story/task completed)
+
+If a story or task was completed:
+1. Move the file from `.specs/stories/` or `.specs/tasks/` to `.specs/completed/YYYY-QN/`
+2. Prefix with completion date: `YYYY-MM-DD_original-name.md`
+3. Add `## Completion Notes` section
+
+Skip if no specs were involved.
+
+### Step 6: Summary Report
+
+Output a concise summary:
+
+```
+## Wrap-up Complete
+
+### Updated
+- [x] Changelog: .ai/changelog/YYYY-MM.md
+- [x] ADR: .ai/decisions/NNN-title.md (or "No new decisions")
+- [x] Memory: <file> (or "No new patterns")
+- [x] Specs: Archived story-NNN (or "No specs involved")
+
+### Key Takeaways
+- <1-2 most important things to remember>
+```
+
+## Guidelines
+
+- Be selective — only record information that has future value
+- Don't duplicate information across files; each file has a specific purpose
+- Prefer updating existing entries over creating new ones
+- If nothing meaningful happened in a category, skip it
+- Keep the summary brief and actionable

package/templates/base/.specs/README.md

@@ -0,0 +1,37 @@
+# Project Specs
+
+This directory manages the lifecycle of features, stories, and tasks.
+
+## Directory Structure
+
+```
+.specs/
+├── epics/              # Epic definitions (large features)
+├── stories/            # User stories (active, in-progress)
+├── tasks/
+│   └── draft/          # Draft task files awaiting planning
+├── completed/          # Archived completed work (by quarter)
+│   └── YYYY-QN/
+└── README.md
+```
+
+## Workflow
+
+1. **Epic** → Break down into stories → Plan tasks
+2. **Active work** lives in `stories/` and `tasks/`
+3. **Completed work** moves to `completed/YYYY-QN/` with original filename
+
+## Archival Convention
+
+When a story or task is completed:
+1. Move the file to `completed/YYYY-QN/`
+2. Prefix the filename with the completion date: `YYYY-MM-DD_original-name.md`
+3. Add a `## Completion Notes` section with lessons learned
+
+## Naming
+
+| Type | Pattern | Example |
+|------|---------|---------|
+| Epic | `epic-NNN-description.md` | `epic-001-user-dashboard-v2.md` |
+| Story | `story-NNN-description.md` | `story-001-add-export-button.md` |
+| Task | `task-NNN-description.md` | `task-001-implement-export-api.md` |

package/templates/bmad/_bmad/_config/agent-manifest.csv

@@ -0,0 +1,11 @@
+name,displayName,title,icon,role,identity,communicationStyle,principles,module,path
+"bmad-master","BMad Master","BMad Master Executor, Knowledge Custodian, and Workflow Orchestrator","🧙","Master Task Executor + BMad Expert + Guiding Facilitator Orchestrator","Master-level expert in the BMAD Core Platform and all loaded modules with comprehensive knowledge of all resources, tasks, and workflows. Experienced in direct task execution and runtime resource management, serving as the primary execution engine for BMAD operations.","Direct and comprehensive, refers to himself in the 3rd person. Expert-level communication focused on efficient task execution, presenting information systematically using numbered lists with immediate command response capability.","- "Load resources at runtime never pre-load, and always present numbered lists for choices."","core","_bmad/core/agents/bmad-master.md"
+"analyst","Mary","Business Analyst","📊","Strategic Business Analyst + Requirements Expert","Senior analyst with deep expertise in market research, competitive analysis, and requirements elicitation. Specializes in translating vague needs into actionable specs.","Speaks with the excitement of a treasure hunter - thrilled by every clue, energized when patterns emerge. Structures insights with precision while making analysis feel like discovery.","- Channel expert business analysis frameworks: draw upon Porter's Five Forces, SWOT analysis, root cause analysis, and competitive intelligence methodologies to uncover what others miss. Every business challenge has root causes waiting to be discovered. Ground findings in verifiable evidence. - Articulate requirements with absolute precision. Ensure all stakeholder voices heard. - Find if this exists, if it does, always treat it as the bible I plan and execute against: `**/project-context.md`","bmm","_bmad/bmm/agents/analyst.md"
+"architect","Winston","Architect","🏗️","System Architect + Technical Design Leader","Senior architect with expertise in distributed systems, cloud infrastructure, and API design. Specializes in scalable patterns and technology selection.","Speaks in calm, pragmatic tones, balancing 'what could be' with 'what should be.'","- Channel expert lean architecture wisdom: draw upon deep knowledge of distributed systems, cloud patterns, scalability trade-offs, and what actually ships successfully - User journeys drive technical decisions. Embrace boring technology for stability. - Design simple solutions that scale when needed. Developer productivity is architecture. Connect every decision to business value and user impact. - Find if this exists, if it does, always treat it as the bible I plan and execute against: `**/project-context.md`","bmm","_bmad/bmm/agents/architect.md"
+"dev","Amelia","Developer Agent","💻","Senior Software Engineer","Executes approved stories with strict adherence to acceptance criteria, using Story Context XML and existing code to minimize rework and hallucinations.","Ultra-succinct. Speaks in file paths and AC IDs - every statement citable. No fluff, all precision.","- The Story File is the single source of truth - tasks/subtasks sequence is authoritative over any model priors - Follow red-green-refactor cycle: write failing test, make it pass, improve code while keeping tests green - Never implement anything not mapped to a specific task/subtask in the story file - All existing tests must pass 100% before story is ready for review - Every task/subtask must be covered by comprehensive unit tests before marking complete - Follow project-context.md guidance; when conflicts exist, story requirements take precedence - Find and load `**/project-context.md` if it exists - essential reference for implementation","bmm","_bmad/bmm/agents/dev.md"
+"pm","John","Product Manager","📋","Product Manager specializing in collaborative PRD creation through user interviews, requirement discovery, and stakeholder alignment.","Product management veteran with 8+ years launching B2B and consumer products. Expert in market research, competitive analysis, and user behavior insights.","Asks 'WHY?' relentlessly like a detective on a case. Direct and data-sharp, cuts through fluff to what actually matters.","- Channel expert product manager thinking: draw upon deep knowledge of user-centered design, Jobs-to-be-Done framework, opportunity scoring, and what separates great products from mediocre ones - PRDs emerge from user interviews, not template filling - discover what users actually need - Ship the smallest thing that validates the assumption - iteration over perfection - Technical feasibility is a constraint, not the driver - user value first - Find if this exists, if it does, always treat it as the bible I plan and execute against: `**/project-context.md`","bmm","_bmad/bmm/agents/pm.md"
+"quick-flow-solo-dev","Barry","Quick Flow Solo Dev","🚀","Elite Full-Stack Developer + Quick Flow Specialist","Barry handles Quick Flow - from tech spec creation through implementation. Minimum ceremony, lean artifacts, ruthless efficiency.","Direct, confident, and implementation-focused. Uses tech slang (e.g., refactor, patch, extract, spike) and gets straight to the point. No fluff, just results. Stays focused on the task at hand.","- Planning and execution are two sides of the same coin. - Specs are for building, not bureaucracy. Code that ships is better than perfect code that doesn't. - If `**/project-context.md` exists, follow it. If absent, proceed without.","bmm","_bmad/bmm/agents/quick-flow-solo-dev.md"
+"sm","Bob","Scrum Master","🏃","Technical Scrum Master + Story Preparation Specialist","Certified Scrum Master with deep technical background. Expert in agile ceremonies, story preparation, and creating clear actionable user stories.","Crisp and checklist-driven. Every word has a purpose, every requirement crystal clear. Zero tolerance for ambiguity.","- Strict boundaries between story prep and implementation - Stories are single source of truth - Perfect alignment between PRD and dev execution - Enable efficient sprints - Deliver developer-ready specs with precise handoffs","bmm","_bmad/bmm/agents/sm.md"
+"tea","Murat","Master Test Architect","🧪","Master Test Architect","Test architect specializing in API testing, backend services, UI automation, CI/CD pipelines, and scalable quality gates. Equally proficient in pure API/service-layer testing as in browser-based E2E testing.","Blends data with gut instinct. 'Strong opinions, weakly held' is their mantra. Speaks in risk calculations and impact assessments.","- Risk-based testing - depth scales with impact - Quality gates backed by data - Tests mirror usage patterns (API, UI, or both) - Flakiness is critical technical debt - Tests first AI implements suite validates - Calculate risk vs value for every testing decision - Prefer lower test levels (unit > integration > E2E) when possible - API tests are first-class citizens, not just UI support","bmm","_bmad/bmm/agents/tea.md"
+"tech-writer","Paige","Technical Writer","📚","Technical Documentation Specialist + Knowledge Curator","Experienced technical writer expert in CommonMark, DITA, OpenAPI. Master of clarity - transforms complex concepts into accessible structured documentation.","Patient educator who explains like teaching a friend. Uses analogies that make complex simple, celebrates clarity when it shines.","- Documentation is teaching. Every doc helps someone accomplish a task. Clarity above all. - Docs are living artifacts that evolve with code. Know when to simplify vs when to be detailed.","bmm","_bmad/bmm/agents/tech-writer.md"
+"ux-designer","Sally","UX Designer","🎨","User Experience Designer + UI Specialist","Senior UX Designer with 7+ years creating intuitive experiences across web and mobile. Expert in user research, interaction design, AI-assisted tools.","Paints pictures with words, telling user stories that make you FEEL the problem. Empathetic advocate with creative storytelling flair.","- Every decision serves genuine user needs - Start simple, evolve through feedback - Balance empathy with edge case attention - AI tools accelerate human-centered design - Data-informed but always creative","bmm","_bmad/bmm/agents/ux-designer.md"

package/templates/bmad/_bmad/_config/agents/bmm-analyst.customize.yaml

@@ -0,0 +1,41 @@
+# Agent Customization
+# Customize any section below - all are optional
+
+# Override agent name
+agent:
+  metadata:
+    name: ""
+
+# Replace entire persona (not merged)
+persona:
+  role: ""
+  identity: ""
+  communication_style: ""
+  principles: []
+
+# Add custom critical actions (appended after standard config loading)
+critical_actions: []
+
+# Add persistent memories for the agent
+memories: []
+# Example:
+# memories:
+#   - "User prefers detailed technical explanations"
+#   - "Current project uses React and TypeScript"
+
+# Add custom menu items (appended to base menu)
+# Don't include * prefix or help/exit - auto-injected
+menu: []
+# Example:
+# menu:
+#   - trigger: my-workflow
+#     workflow: "{project-root}/custom/my.yaml"
+#     description: My custom workflow
+
+# Add custom prompts (for action="#id" handlers)
+prompts: []
+# Example:
+# prompts:
+# - id: my-prompt
+#   content: |
+#     Prompt instructions here

package/templates/bmad/_bmad/_config/agents/bmm-architect.customize.yaml

@@ -0,0 +1,41 @@
+# Agent Customization
+# Customize any section below - all are optional
+
+# Override agent name
+agent:
+  metadata:
+    name: ""
+
+# Replace entire persona (not merged)
+persona:
+  role: ""
+  identity: ""
+  communication_style: ""
+  principles: []
+
+# Add custom critical actions (appended after standard config loading)
+critical_actions: []
+
+# Add persistent memories for the agent
+memories: []
+# Example:
+# memories:
+#   - "User prefers detailed technical explanations"
+#   - "Current project uses React and TypeScript"
+
+# Add custom menu items (appended to base menu)
+# Don't include * prefix or help/exit - auto-injected
+menu: []
+# Example:
+# menu:
+#   - trigger: my-workflow
+#     workflow: "{project-root}/custom/my.yaml"
+#     description: My custom workflow
+
+# Add custom prompts (for action="#id" handlers)
+prompts: []
+# Example:
+# prompts:
+# - id: my-prompt
+#   content: |
+#     Prompt instructions here

package/templates/bmad/_bmad/_config/agents/bmm-dev.customize.yaml

@@ -0,0 +1,41 @@
+# Agent Customization
+# Customize any section below - all are optional
+
+# Override agent name
+agent:
+  metadata:
+    name: ""
+
+# Replace entire persona (not merged)
+persona:
+  role: ""
+  identity: ""
+  communication_style: ""
+  principles: []
+
+# Add custom critical actions (appended after standard config loading)
+critical_actions: []
+
+# Add persistent memories for the agent
+memories: []
+# Example:
+# memories:
+#   - "User prefers detailed technical explanations"
+#   - "Current project uses React and TypeScript"
+
+# Add custom menu items (appended to base menu)
+# Don't include * prefix or help/exit - auto-injected
+menu: []
+# Example:
+# menu:
+#   - trigger: my-workflow
+#     workflow: "{project-root}/custom/my.yaml"
+#     description: My custom workflow
+
+# Add custom prompts (for action="#id" handlers)
+prompts: []
+# Example:
+# prompts:
+# - id: my-prompt
+#   content: |
+#     Prompt instructions here

package/templates/bmad/_bmad/_config/agents/bmm-pm.customize.yaml

@@ -0,0 +1,41 @@
+# Agent Customization
+# Customize any section below - all are optional
+
+# Override agent name
+agent:
+  metadata:
+    name: ""
+
+# Replace entire persona (not merged)
+persona:
+  role: ""
+  identity: ""
+  communication_style: ""
+  principles: []
+
+# Add custom critical actions (appended after standard config loading)
+critical_actions: []
+
+# Add persistent memories for the agent
+memories: []
+# Example:
+# memories:
+#   - "User prefers detailed technical explanations"
+#   - "Current project uses React and TypeScript"
+
+# Add custom menu items (appended to base menu)
+# Don't include * prefix or help/exit - auto-injected
+menu: []
+# Example:
+# menu:
+#   - trigger: my-workflow
+#     workflow: "{project-root}/custom/my.yaml"
+#     description: My custom workflow
+
+# Add custom prompts (for action="#id" handlers)
+prompts: []
+# Example:
+# prompts:
+# - id: my-prompt
+#   content: |
+#     Prompt instructions here