Coordinate multiple agents for complex tasks. Use for multi-perspective analysis, comprehensive reviews, or tasks requiring different domain expertise.
Coordinate multiple agents for complex tasks. Use for multi-perspective analysis, comprehensive reviews, or tasks requiring different domain expertise.
Multi-Agent Orchestration
You are now in ORCHESTRATION MODE. Your task: coordinate specialized agents to solve this complex problem.
Task to Orchestrate
$ARGUMENTS
🔴 CRITICAL: Minimum Agent Requirement
⚠️ ORCHESTRATION = MINIMUM 3 DIFFERENT AGENTS
If you use fewer than 3 agents, you are NOT orchestrating - you're just delegating.
Use the project-planner agent to create PLAN.md
→ STOP after plan is created
→ ASK user for approval
PHASE 2 (Implementation - after approval):
Invoke agents in PARALLEL:
Use the frontend-specialist agent to [task]
Use the backend-specialist agent to [task]
Use the test-engineer agent to [task]
🔴 CRITICAL: Context Passing (MANDATORY)
When invoking ANY subagent, you MUST include:
Original User Request: Full text of what user asked
Decisions Made: All user answers to Socratic questions
Previous Agent Work: Summary of what previous agents did
Current Plan State: If plan files exist in workspace, include them
Example with FULL context:
Use the project-planner agent to create PLAN.md:
**CONTEXT:**
- User Request: "A social platform for students, using mock data"
- Decisions: Tech=Vue 3, Layout=Grid Widgets, Auth=Mock, Design=Youthful & dynamic
- Previous Work: Orchestrator asked 6 questions, user chose all options
- Current Plan: playful-roaming-dream.md exists in workspace with initial structure
**TASK:** Create detailed PLAN.md based on ABOVE decisions. Do NOT infer from folder name.
⚠️ VIOLATION: Invoking subagent without full context = subagent will make wrong assumptions!
Step 4: Verification (MANDATORY)
The LAST agent must run appropriate verification scripts:
