squad-station 0.6.7 → 0.6.8

package/.squad/examples/orchestrator-claude.yml

@@ -10,6 +10,8 @@ sdd:
   #   playbook: ".squad/sdd/bmad-playbook.md"
   # - name: superpowers
   #   playbook: ".squad/sdd/superpowers-playbook.md"
+  # - name: openspec
+  #   playbook: ".squad/sdd/openspec-playbook.md"
 
 orchestrator:
   provider: claude-code

package/.squad/examples/orchestrator-gemini.yml

@@ -10,6 +10,8 @@ sdd:
   #   playbook: ".squad/sdd/bmad-playbook.md"
   # - name: superpowers
   #   playbook: ".squad/sdd/superpowers-playbook.md"
+  # - name: openspec
+  #   playbook: ".squad/sdd/openspec-playbook.md"
 
 orchestrator:
   provider: gemini-cli

package/.squad/sdd/bmad-playbook.md

@@ -1,429 +1,86 @@
-# BMad Method — Playbook v6.2.0
+# BMad Method — Agent Playbook
 
-> Comprehensive operational guide for BMad Method — installation, Day 1 workflow, cheat sheet, troubleshooting, best practices.
+## Workflow Sequence
 
----
+**Greenfield (new project):**
+1. `/bmad-brainstorming` — Explore ideas
+2. `/bmad-product-brief-preview` — Create product brief (guided/yolo/autonomous)
+3. `/bmad-create-prd` — Write requirements
+4. `/bmad-create-ux-design` — UX design document
+5. `/bmad-create-architecture` — Architecture document
+6. `/bmad-create-epics-and-stories` — Break PRD into epics
+7. `/bmad-check-implementation-readiness` — Validate planning cohesion
+8. `/bmad-sprint-planning` — Initialize sprint tracking
+9. **Build Cycle** (repeat per story): `create-story` → `dev-story` → `code-review` → `retrospective`
 
-## Installation & Setup
+**Brownfield (existing project):**
+1. `/bmad-help` — Detect project state, get recommendations
+2. `/bmad-generate-project-context` — Document tech preferences
+3. Pick appropriate phase above based on project state
 
-### Prerequisites
+**Quick fix / small change:** `/bmad-quick-dev` — Unified workflow (clarify → plan → implement → review → present)
 
-| Requirement | Version | Notes |
-|---|---|---|
-| **Node.js** | ≥ 20.0.0 | Required |
-| **Git** | Latest | Recommended |
-| **AI IDE** | Latest | Claude Code, Cursor, Gemini CLI, Codex CLI, +16 other platforms |
-| **npm** | Bundled with Node.js | Needed for `npx` |
+## Command Reference
 
-### Installation — Interactive Mode
+### Phase 1: Analysis
 
-```bash
-# Install BMad Method (interactive — stable version)
-npx bmad-method install
-
-# Get the latest from tip of main (prerelease)
-npx bmad-method@next install
-
-# Specify exact version if stuck on stale cache
-npx bmad-method@6.2.0 install
-```
-
-**Steps the installer will ask:**
-
-1. **Select modules** — Choose modules to install (BMM, BMB, TEA, GDS, CIS, WDS)
-2. **Select IDE** — Choose your AI IDE (20+ platforms: Claude Code, Cursor, Gemini CLI...)
-3. **Configure paths** — Output artifacts path
-4. **Project name** — Project name
-5. **Skill level** — beginner / intermediate / expert
-6. **Confirm** — Review and confirm
-
-### Installation — Non-Interactive Mode (CI/CD)
-
-```bash
-# Basic installation
-npx bmad-method install --directory /path/to/project --modules bmm --tools claude-code --yes
-
-# Install multiple modules
-npx bmad-method install --directory ./my-project --modules bmm,tea,bmb --tools cursor --yes
-
-# With custom content
-npx bmad-method install --directory ./my-project --modules bmm --tools claude-code --custom-content ./custom --yes
-```
-
-| Flag | Description |
+| Command | Description |
 |---|---|
-| `--directory <path>` | Project path |
-| `--modules <list>` | Modules to install (comma-separated: `bmm`, `bmb`, `tea`, `gds`, `cis`, `wds`) |
-| `--tools <ide>` | IDE target: `claude-code`, `cursor`, `gemini`, `codex`, `windsurf`, `opencode`, +14 others |
-| `--yes` | Skip confirmation |
-| `--custom-content <path>` | Custom content directory |
-
-### Post-Installation Result
-
-```
-your-project/
-├── _bmad/                         # BMad configuration
-│   ├── bmm/                       # BMM module
-│   │   ├── config.yaml            # Resolved settings
-│   │   ├── agents/                # Agent persona skill dirs
-│   │   ├── skills/                # Workflow skill dirs
-│   │   └── data/                  # Reference data
-│   └── core/                      # Core skills
-├── _bmad-output/
-│   ├── planning-artifacts/        # PRD, Architecture, Epics
-│   ├── implementation-artifacts/  # Sprint status, Stories
-│   └── project-context.md         # (optional) Implementation rules
-├── .claude/skills/                # Installed skills (depends on IDE)
-│   ├── bmad-help/SKILL.md
-│   ├── bmad-create-prd/SKILL.md
-│   └── ...                        # 44 SKILL.md files total
-└── docs/                          # Project knowledge
-```
-
-### Uninstall
-
-```bash
-# Interactive uninstall
-npx bmad-method uninstall
-
-# Or remove manually
-rm -rf _bmad/ _bmad-output/ .claude/skills/bmad-*
-```
-
-### Upgrade from 6.0.x → 6.2
-
-```bash
-# 1. Backup current artifacts
-cp -r _bmad-output/ _bmad-output-backup/
-
-# 2. Re-run installer
-npx bmad-method@6.2.0 install
-
-# 3. Verify skills have regenerated
-ls .claude/skills/
-```
-
-> ⚠️ **Breaking changes V6.1:** The entire system has migrated to **skills-based architecture**. Legacy YAML/XML workflow engine has been completely removed. Path variables changed from `@` prefix to `{project-root}` syntax. Terminology: "commands" → "skills".
-
----
-
-## Day 1 Workflow
-
-### Greenfield Project (New Project)
-
-```mermaid
-flowchart TD
-    A["Install BMad"] --> B["Open AI IDE"]
-    B --> C["Run bmad-help"]
-    C --> D{"Choose Planning Track"}
-    D -->|"Bug fix"| QF["Quick Flow"]
-    D -->|"Product"| BM["BMad Method"]
-    D -->|"Enterprise"| ENT["Enterprise"]
-
-    QF --> QD["bmad-quick-dev<br/>(unified workflow)"]
-    QD --> DONE["Done!"]
-
-    BM --> BS["bmad-brainstorming"]
-    BS --> PBR["bmad-product-brief-preview<br/>(NEW — guided/yolo/autonomous)"]
-    PBR --> PRD["bmad-create-prd"]
-    PRD --> UX["bmad-create-ux-design"]
-    UX --> ARCH["bmad-create-architecture"]
-    ARCH --> EPC["bmad-create-epics-and-stories"]
-    EPC --> IRC["bmad-check-implementation-readiness"]
-    IRC --> SP["bmad-sprint-planning"]
-    SP --> BUILD["Build Cycle"]
-
-    ENT --> BS
-```
-
-### Brownfield Project (Existing Project)
-
-1. **Install BMad** — `npx bmad-method install` in the existing project folder
-2. **Run `/bmad-help`** — BMad-Help will detect project state
-3. **Create Project Context** — `/bmad-generate-project-context` to document technical preferences
-4. **Choose track** — Quick Flow for small changes, BMad Method for major features
-5. **Start from the appropriate phase** — Can skip Phase 1 if direction is already clear
-
-### Build Cycle (Repeat for each story)
-
-| Step | Agent | Skill | Command | Description |
-|---|---|---|---|---|
-| 1 | Scrum Master | `bmad-create-story` | `/bmad-create-story` | Create story file from epic |
-| 2 | Developer | `bmad-dev-story` | `/bmad-dev-story` | Implement story (new chat!) |
-| 3 | Developer | `bmad-code-review` | `/bmad-code-review` | Review code (sharded: 4-step parallel review) |
-| 4 | Scrum Master | `bmad-retrospective` | `/bmad-retrospective` | After completing an epic |
-
-> ⚡ **IMPORTANT:** Always use a **fresh chat** for each workflow!
-
----
-
-## Daily Operations
-
-### Daily Operations Checklist
-
-```markdown
-## Every day when starting work:
-1. Open AI IDE in project folder
-2. Run `/bmad-help` → view project state and next steps
-3. Run `/bmad-create-story` for the next story (new chat)
-4. Run `/bmad-dev-story` to implement (new chat)
-5. Run `/bmad-code-review` when finished (new chat)
-6. Commit code frequently
-
-## When an epic is completed:
-1. Run `/bmad-retrospective` (new chat)
-2. Review sprint-status.yaml
-3. Continue to the next epic
-
-## When scope changes:
-1. Run `/bmad-correct-course` with SM agent (new chat)
-2. Update PRD if needed
-3. Re-plan epics
-```
-
-### Creating Project Context (Recommended)
-
-Project Context helps AI agents understand your technical preferences:
-
-```bash
-# Method 1: Auto-generate after having architecture
-/bmad-generate-project-context
-
-# Method 2: Create manually
-# Create file _bmad-output/project-context.md with content:
-```
+| `/bmad-brainstorming` | Guided ideation session |
+| `/bmad-product-brief-preview` | Guided/yolo/autonomous product brief |
+| `/bmad-create-product-brief` | Foundation document |
+| `/bmad-market-research` | Market analysis, competitive landscape |
+| `/bmad-domain-research` | Industry domain deep dive |
+| `/bmad-technical-research` | Technical feasibility analysis |
+| `/bmad-document-project` | Analyze existing project for docs |
 
-```markdown
-# Project Context
+### Phase 2: Planning
 
-## Technology Stack
-- Frontend: React 18 + TypeScript
-- Backend: Node.js + Express
-- Database: PostgreSQL
-- ORM: Prisma
-
-## Implementation Rules
-- Use TypeScript strict mode
-- All API responses must follow REST conventions
-- Error handling with custom error classes
-- Unit test coverage minimum 80%
-```
-
----
-
-## Strategic Configuration
-
-### Presets by Project Type
-
-| Project Type | Modules | Skill Level | Track |
-|---|---|---|---|
-| **Side project / MVP** | `bmm` | intermediate | Quick Flow |
-| **Startup product** | `bmm` + `cis` | intermediate | BMad Method |
-| **Enterprise app** | `bmm` + `tea` | expert | Enterprise |
-| **Game** | `bmm` + `gds` | intermediate | BMad Method |
-| **Custom agents** | `bmm` + `bmb` | expert | BMad Method |
-| **Design-heavy** | `bmm` + `wds` | intermediate | BMad Method |
-| **Full suite** | `bmm` + `bmb` + `tea` + `cis` | expert | Enterprise |
-
-### When to Use Each Agent?
-
-| Situation | Agent | Skill |
-|---|---|---|
-| Have a new idea to explore | Analyst (Mary) | `/bmad-brainstorming` |
-| Need market research | Analyst (Mary) | `/bmad-market-research` |
-| Need domain research | Analyst (Mary) | `/bmad-domain-research` |
-| Need technical research | Analyst (Mary) | `/bmad-technical-research` |
-| Create product brief (new) | Analyst (Mary) | `/bmad-product-brief-preview` |
-| Write requirements | PM (John) | `/bmad-create-prd` |
-| Validate PRD | PM (John) | `/bmad-validate-prd` |
-| Design UI/UX | UX Designer (Sally) | `/bmad-create-ux-design` |
-| Design architecture | Architect (Winston) | `/bmad-create-architecture` |
-| Sprint planning | Scrum Master (Bob) | `/bmad-sprint-planning` |
-| View sprint status | Scrum Master (Bob) | `/bmad-sprint-status` |
-| Implement code | Developer (Amelia) | `/bmad-dev-story` |
-| Review code | Developer (Amelia) | `/bmad-code-review` |
-| E2E tests | QA (Quinn) | `/bmad-qa-generate-e2e-tests` |
-| Bug fix / small change | Quick Flow (Barry) | `/bmad-quick-dev` |
-| Group discussion | Party Mode | `/bmad-party-mode` |
-| Don't know what to do next | BMad-Help | `/bmad-help` |
-
----
-
-## Cheat Sheet
-
-### Core Skills (available in all modules)
-
-| Skill | Phase | Description |
-|---|---|---|
-| `/bmad-help` | Any | 🌟 Intelligent guide — ask anything |
-| `/bmad-brainstorming` | Any | Guided ideation session |
-| `/bmad-party-mode` | Any | Multi-agent collaboration |
-| `/bmad-review-adversarial-general` | Any | Adversarial content review |
-| `/bmad-review-edge-case-hunter` | Any | Edge case analysis for code |
-| `/bmad-distillator` | Any | Lossless LLM-optimized document compression |
-| `/bmad-editorial-review-prose` | Any | Review prose quality |
-| `/bmad-editorial-review-structure` | Any | Review document structure |
-| `/bmad-advanced-elicitation` | Any | Advanced questioning techniques |
-| `/bmad-index-docs` | Any | Create doc index for LLM scanning |
-| `/bmad-shard-doc` | Any | Split large markdown file |
-
-### BMM Phase 1: Analysis Skills
-
-| Skill | Agent | Description |
-|---|---|---|
-| `/bmad-product-brief-preview` | Analyst | 🆕 Guided/Yolo/Autonomous product brief creation |
-| `/bmad-create-product-brief` | Analyst | Foundation document |
-| `/bmad-market-research` | Analyst | Market analysis, competitive landscape |
-| `/bmad-domain-research` | Analyst | Industry domain deep dive |
-| `/bmad-technical-research` | Analyst | Technical feasibility analysis |
-| `/bmad-document-project` | Analyst | Analyze existing project for docs |
-
-### BMM Phase 2: Planning Skills
-
-| Skill | Agent | Description |
-|---|---|---|
-| `/bmad-create-prd` | PM | Product Requirements Document |
-| `/bmad-validate-prd` | PM | Validate PRD completeness |
-| `/bmad-edit-prd` | PM | Improve existing PRD |
-| `/bmad-create-ux-design` | UX | UX Design document |
-
-### BMM Phase 3: Solutioning Skills
-
-| Skill | Agent | Description |
-|---|---|---|
-| `/bmad-create-architecture` | Architect | Architecture document |
-| `/bmad-create-epics-and-stories` | PM | Break PRD → epics |
-| `/bmad-check-implementation-readiness` | Architect | Validate planning cohesion |
-| `/bmad-generate-project-context` | Analyst | Generate project context from codebase |
-
-### BMM Phase 4: Implementation Skills
-
-| Skill | Agent | Description |
-|---|---|---|
-| `/bmad-sprint-planning` | SM | Initialize sprint tracking |
-| `/bmad-sprint-status` | SM | 🆕 Summarize sprint status |
-| `/bmad-create-story` | SM | Create story file |
-| `/bmad-dev-story` | Developer | Implement story |
-| `/bmad-quick-dev` | Barry | 🆕 Unified quick flow (clarify → plan → implement → review → present) |
-| `/bmad-code-review` | Developer | 🆕 Sharded parallel code review (4 steps) |
-| `/bmad-qa-generate-e2e-tests` | QA | Generate E2E tests |
-| `/bmad-correct-course` | SM | Handle scope changes |
-| `/bmad-retrospective` | SM | Epic retrospective |
-
-### Agent Skills (Load persona)
-
-| Skill | Agent Persona |
+| Command | Description |
 |---|---|
-| `/bmad-agent-analyst` | Mary — Analyst |
-| `/bmad-agent-pm` | John — Product Manager |
-| `/bmad-agent-architect` | Winston — Architect |
-| `/bmad-agent-sm` | Bob — Scrum Master |
-| `/bmad-agent-dev` | Amelia — Developer |
-| `/bmad-agent-qa` | Quinn — QA Engineer |
-| `/bmad-agent-quick-flow-solo-dev` | Barry — Quick Flow Solo Dev |
-| `/bmad-agent-ux-designer` | Sally — UX Designer |
-| `/bmad-agent-tech-writer` | Paige — Technical Writer |
+| `/bmad-create-prd` | Product Requirements Document |
+| `/bmad-validate-prd` | Validate PRD completeness |
+| `/bmad-edit-prd` | Improve existing PRD |
+| `/bmad-create-ux-design` | UX Design document |
 
-### CLI Commands
+### Phase 3: Solutioning
 
 | Command | Description |
 |---|---|
-| `npx bmad-method install` | Interactive install |
-| `npx bmad-method@next install` | Install latest prerelease |
-| `npx bmad-method install --directory <path> --modules <m> --tools <ide> --yes` | Non-interactive |
-| `npx bmad-method uninstall` | Remove components |
-| `npx bmad-method@<version> install` | Install specific version |
-
----
-
-## Troubleshooting
-
-| Error | Cause | Fix |
-|---|---|---|
-| **Skills not showing in IDE** | IDE hasn't enabled skills or needs restart | Check IDE settings → enable skills → restart/reload |
-| **`npx bmad-method` gives old version** | npm cache is stale | `npx bmad-method@6.2.0 install` or `@next` |
-| **Old module skills still present** | Installer doesn't auto-delete | Remove manually: `rm -rf .claude/skills/bmad-*` → re-install |
-| **Nested install rejected** | Ancestor directory already has BMAD | Install at project root, no nesting |
-| **Workflow context errors** | Running multiple workflows in same chat | **Always use fresh chat** for each workflow |
-| **Brainstorming overwrites old session** | Bug v6.0.3 and below | Update to v6.0.4+ |
-| **Agent responds with wrong role** | Context noise from previous workflow | Fresh chat + invoke correct agent skill |
-| **Quick Dev scope creep** | Feature larger than expected | Quick Dev auto-detect → split or escalate |
-| **Code Review infinite loop** | Old bug: mandatory minimum findings | Fixed v6.1+ — removed mandatory minimum |
-| **PRD loses brainstorming ideas** | Silent loss bug | Fixed v6.1+ — added reconciliation step |
-| **Cannot find module** | Node.js version < 20 | Upgrade Node.js: `nvm install 20` |
-| **Skill validation errors** | Naming/path/variable issues | Run skill-validator.md on skill directory |
-
-### Debug Steps
-
-```bash
-# 1. Verify Node.js version
-node --version  # Must be >= 20
-
-# 2. Clear npm cache
-npm cache clean --force
-
-# 3. Re-install with specific version or @next
-npx bmad-method@6.2.0 install
-# or
-npx bmad-method@next install
-
-# 4. Verify skills directory
-ls -la .claude/skills/  # or .cursor/skills/
-
-# 5. Check installed config
-cat _bmad/bmm/config.yaml
+| `/bmad-create-architecture` | Architecture document |
+| `/bmad-create-epics-and-stories` | Break PRD into epics |
+| `/bmad-check-implementation-readiness` | Validate cohesion between PRD, Architecture, Epics |
+| `/bmad-generate-project-context` | Generate project context from codebase |
 
-# 6. Report bug if still failing
-# https://github.com/bmad-code-org/BMAD-METHOD/issues
-```
+### Phase 4: Implementation
 
----
-
-## Best Practices
-
-### ✅ Gold Rules
-
-| # | Rule | Why |
-|---|---|---|
-| 1 | **Always use fresh chat** for each workflow | Avoid context pollution, each agent needs clean context |
-| 2 | **Start with `/bmad-help`** | BMad-Help inspects project state → recommends exact next step |
-| 3 | **Create Project Context early** | Agents understand tech preferences → higher quality output |
-| 4 | **Choose the right planning track** | Quick Flow for small, BMad Method for medium, Enterprise for complex |
-| 5 | **Run Implementation Readiness** before coding | Validate cohesion between PRD, Architecture, Epics |
-| 6 | **Code Review every story** | Sharded parallel review catches bugs early (Blind Hunter + Edge Case Hunter + Acceptance Auditor) |
-| 7 | **Retrospective every epic** | Continuous improvement, adjust sprint |
-| 8 | **Commit code frequently** | Keep diffs small, easy to review |
-| 9 | **Git ignore planning artifacts** if containing sensitive info | PRD/Architecture may contain business logic |
-| 10 | **Upgrade BMad regularly** | Bug fixes, new features, better agent behavior. Use `@next` for cutting edge |
-
-### ❌ Anti-Patterns
-
-| Anti-Pattern | Problem | Solution |
-|---|---|---|
-| **Running multiple workflows in same chat** | Context noise, agent confused | Fresh chat for each workflow |
-| **Skip PRD and go straight to code** (for medium+ projects) | Missing requirements → excessive rework | At minimum need PRD or tech-spec |
-| **Vibe-coding instead of following process** | Inconsistent output | Use structured workflows |
-| **Ignore `bmad-help` recommendations** | Miss important steps | Trust bmad-help, it inspects project state |
-| **Not creating Project Context** | Agents don't understand conventions | Create `project-context.md` early |
-| **Nested BMAD install** | Duplicate commands, confusion | Install at root project only |
-| **Using Quick Flow for complex features** | Under-planned, scope creep | Escalate to BMad Method when scope is large |
-| **Not running Code Review** | Bugs slip through | Always `/bmad-code-review` after every story |
-| **Reference files across skills** | Breaking encapsulation (PATH-05) | Use `skill:skill-name` syntax instead of file paths |
-
----
-
-## Resources
+| Command | Description |
+|---|---|
+| `/bmad-sprint-planning` | Initialize sprint tracking |
+| `/bmad-sprint-status` | Summarize sprint status |
+| `/bmad-create-story` | Create story file from epic |
+| `/bmad-dev-story` | Implement story |
+| `/bmad-quick-dev` | Unified quick flow |
+| `/bmad-code-review` | Sharded parallel code review (4 steps) |
+| `/bmad-qa-generate-e2e-tests` | Generate E2E tests |
+| `/bmad-correct-course` | Handle scope changes |
+| `/bmad-retrospective` | Epic retrospective |
+
+### Utilities
 
-| Resource | Link |
+| Command | Description |
 |---|---|
-| **Documentation** | [docs.bmad-method.org](https://docs.bmad-method.org) |
-| **Getting Started** | [Tutorial](https://docs.bmad-method.org/tutorials/getting-started/) |
-| **Upgrade to V6** | [Guide](https://docs.bmad-method.org/how-to/upgrade-to-v6/) |
-| **GitHub** | [bmad-code-org/BMAD-METHOD](https://github.com/bmad-code-org/BMAD-METHOD) |
-| **Discord** | [Join Community](https://discord.gg/gk8jAdXWmj) |
-| **YouTube** | [BMadCode Channel](https://www.youtube.com/@BMadCode) |
-| **npm** | [bmad-method](https://www.npmjs.com/package/bmad-method) |
-| **Issues** | [GitHub Issues](https://github.com/bmad-code-org/BMAD-METHOD/issues) |
-| **Discussions** | [GitHub Discussions](https://github.com/bmad-code-org/BMAD-METHOD/discussions) |
-| **Roadmap** | [docs.bmad-method.org/roadmap](https://docs.bmad-method.org/roadmap/) |
-| **Chinese Docs** | [README_CN.md](https://github.com/bmad-code-org/BMAD-METHOD/blob/main/README_CN.md) |
+| `/bmad-help` | Intelligent guide — ask anything |
+| `/bmad-party-mode` | Multi-agent collaboration |
+| `/bmad-review-adversarial-general` | Adversarial content review |
+| `/bmad-review-edge-case-hunter` | Edge case analysis for code |
+| `/bmad-distillator` | Lossless LLM-optimized document compression |
+| `/bmad-index-docs` | Create doc index for LLM scanning |
+| `/bmad-shard-doc` | Split large markdown file |
+
+## Critical Rules
+
+1. **Fresh chat per workflow** — Each workflow must run in a clean context. Never chain workflows.
+2. **Run implementation-readiness before coding** — Validates cohesion between PRD, Architecture, and Epics.
+3. **Code review every story** — Always `/bmad-code-review` after `/bmad-dev-story`.
+4. **Start with `/bmad-help`** — It inspects project state and recommends the exact next step.