Agent Orchestration — coordinate multiple specialized AI agents
Design systems where a planner agent decomposes tasks, specialized agents execute subtasks, and a synthesizer agent combines results.
Agent4.3
Prompt template
You are the Orchestrator. Coordinate the following specialists:
- Researcher: gathers and verifies information
- Writer: drafts and polishes content
- Critic: reviews and identifies gaps
- Coder: writes and tests code
Task: {task}
1. Decompose into subtasks
2. Assign each to a specialist
3. Review outputs and request revisions
4. Synthesize final resultHow to use
- 1Define specialist roles with clear capabilities and limits
- 2Planner agent decomposes task into ordered subtasks
- 3Route each subtask to the appropriate specialist
- 4Critic agent reviews each output before passing on
- 5Synthesizer combines all outputs into final deliverable
Pro tips
- Use a shared 'scratchpad' for inter-agent communication
- Limit each agent's context to only what it needs
- Add a 'terminator' agent that decides when the task is done
- Monitor token cost — multi-agent systems can be 3-5x more expensive
Multi-AgentOrchestrationPlannerSpecialistclaude
Ratings & Reviews
Loading reviews…
Related Skills
Agent
ReAct Prompting — reasoning + acting for tool-using agents
Alternate between Thought (reasoning) and Action (tool call) steps so the AI can solve multi-step tasks with external tools.
ReActAgentTool Use
4.6
Agent
Reflection Agents — self-critique loops that improve output quality
Have the agent generate a draft, critique its own work against explicit criteria, then revise — repeating until quality thresholds are met.
AgentReflectionSelf-Critique
4.3
Agent
Plan-and-Execute — separate planning from execution for complex tasks
First create a detailed step-by-step plan, then execute each step independently with tool calls, re-planning if an execution step fails.
AgentPlanningExecution
4.5