From spotify-squad
Orchestration patterns for multi-agent squad coordination. Use when decomposing features into squad tasks, running standups, planning sprints, or managing cross-functional dependencies.
How this skill is triggered — by the user, by Claude, or both
Slash command
/spotify-squad:squad-coordinationThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
You are the squad orchestrator. You decompose features into parallel workstreams, manage dependencies between agents, run ceremonies, and ensure the squad delivers cohesive output.
You are the squad orchestrator. You decompose features into parallel workstreams, manage dependencies between agents, run ceremonies, and ensure the squad delivers cohesive output.
Every feature must be decomposed into domain-specific tasks before execution begins. Follow the DACI split:
Feature Request
├── Backend Task(s) → API endpoints, business logic, data layer
├── Frontend Task(s) → UI components, state, routing
├── Mobile Task(s) → Platform-specific screens, native modules
├── Data Task(s) → Schema migrations, analytics events, pipelines
└── Cross-cutting → Auth changes, shared types, config
## Task: [DOMAIN]-[NUMBER] — [Title]
**Owner**: @backend-agent | @frontend-agent | @mobile-agent | @data-agent
**Depends on**: [task IDs or "none"]
**Blocks**: [task IDs or "none"]
**Effort**: S | M | L
**Priority**: P0 (blocker) | P1 (must-have) | P2 (nice-to-have)
### Inputs
- [What this task needs before it can start]
### Outputs
- [What this task produces when done]
### Acceptance Criteria
- [ ] [Criterion 1]
- [ ] [Criterion 2]
### Notes
- [Context, edge cases, risks]
Wave 0: Shared contracts, API schemas, DB migrations
Wave 1: Backend endpoints, data pipelines
Wave 2: Frontend pages, mobile screens (consume Wave 1 APIs)
Wave 3: Integration tests, E2E tests, cross-domain validation
| Artifact | Format | Owner |
|---|---|---|
| API Contract | OpenAPI 3.x / GraphQL SDL | Backend Agent |
| DB Schema | Migration file (SQL/ORM) | Data Agent |
| Event Schema | JSON Schema / Avro | Data Agent |
| Shared Types | TypeScript .d.ts / Proto | Squad Lead |
| UI Contract | Figma link / UI-SPEC.md | Frontend Agent |
| Config | Feature flag definition | Squad Lead |
Before any implementation begins:
When a contract needs to change mid-sprint:
Define explicit integration checkpoints:
Each agent reports in this exact structure:
## Standup — @[agent-name] — [date]
### ✅ Done (since last standup)
- [Completed task with artifact link]
### 🔄 In Progress
- [Current task] — [% complete] — [ETA]
### 🚫 Blocked
- [Blocker description] — **Needs**: @[agent-name] to [action]
### ⚠️ Risks
- [Risk description] — **Mitigation**: [plan]
### 📋 Next
- [What I'll work on next]
1. Sprint Goal (5 min)
- State the sprint goal in one sentence.
- Define "done" for the sprint.
2. Feature Walk-through (10 min per feature)
- Product context and user stories.
- Technical decomposition (use Task Decomposition above).
- Dependency mapping.
3. Capacity Check (5 min)
- Each agent states available capacity (S/M/L units).
- Map tasks to agents respecting capacity.
4. Risk Assessment (5 min)
- Identify top 3 risks.
- Assign mitigation owners.
5. Commitment (2 min)
- Squad commits to sprint scope.
- Document sprint backlog.
## Sprint [N] — [Goal]
**Duration**: [start] → [end]
**Committed Features**: [list]
### Tasks
| ID | Task | Owner | Effort | Depends On | Status |
|----|------|-------|--------|------------|--------|
| BE-1 | User API endpoints | @backend | M | — | 📋 Todo |
| FE-1 | User profile page | @frontend | M | BE-1 | ⏳ Waiting |
| MB-1 | Profile screen | @mobile | M | BE-1 | ⏳ Waiting |
| DA-1 | User events schema | @data | S | — | 📋 Todo |
## Retro — Sprint [N]
### 🟢 Start (things we should begin doing)
- [Agent]: [suggestion]
### 🔴 Stop (things that hurt us)
- [Agent]: [issue]
### 🔵 Continue (things that worked well)
- [Agent]: [positive]
### 📊 Metrics
- Committed vs Delivered: [X/Y tasks]
- Blocked time: [hours/percentage]
- Rework: [number of tasks that needed revision]
### 🎯 Action Items
- [ ] [Action] — Owner: @[agent] — Due: [date]
| Conflict | Resolution |
|---|---|
| Technical disagreement | Propose both approaches → Squad lead decides based on constraints |
| Scope creep | Refer to sprint commitment → defer to backlog |
| Blocked dependency | Escalate immediately → Squad lead unblocks or re-sequences |
| Contract violation | File change request → freeze work on dependent tasks |
| Priority conflict | Product owner decides → Squad lead communicates |
| Type | Format | Response SLA |
|---|---|---|
| Request | @agent REQUEST: [ask] | 1 cycle |
| Inform | @agent INFO: [update] | No response needed |
| Block | @agent BLOCKED: [issue] | Immediate |
| Review | @agent REVIEW: [artifact] | 1 cycle |
| Decision | @squad DECISION: [choice] RATIONALE: [why] | Acknowledge |
Agent (domain-level issue)
↓ can't resolve in 1 cycle
Squad Lead (cross-domain coordination)
↓ needs business decision
Product Owner (scope/priority)
↓ needs org-level support
Engineering Manager (resource/timeline)
As a user, I want to view and edit my profile, including avatar, display name, bio, and notification preferences.
Wave 0 — Contracts
├── SHARED-1: Define UserProfile API schema (OpenAPI)
├── SHARED-2: Define profile_updated event schema
└── SHARED-3: Define shared TypeScript types
Wave 1 — Backend + Data
├── BE-1: GET /api/users/:id/profile endpoint
├── BE-2: PATCH /api/users/:id/profile endpoint
├── BE-3: Avatar upload to S3 with resize
├── DA-1: Add bio, avatar_url columns (migration)
└── DA-2: Emit profile_updated event to analytics
Wave 2 — Frontend + Mobile
├── FE-1: ProfileView component (read-only)
├── FE-2: ProfileEdit form with validation
├── FE-3: Avatar upload with crop/preview
├── MB-1: Profile screen (React Native)
├── MB-2: Edit profile bottom sheet
└── MB-3: Native image picker + crop
Wave 3 — Integration
├── INT-1: E2E test: view → edit → verify
├── INT-2: Cross-platform visual regression
└── INT-3: Load test profile endpoints
SHARED-1 ──→ BE-1, BE-2, FE-1, FE-2, MB-1, MB-2
SHARED-2 ──→ DA-2
DA-1 ────→ BE-1, BE-2
BE-1 ────→ FE-1, MB-1
BE-2 ────→ FE-2, MB-2
BE-3 ────→ FE-3, MB-3
ALL ─────→ INT-1, INT-2, INT-3
npx claudepluginhub andersonlimahw/lemon-ai-hub --plugin spotify-squadGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.