The Idea
A single AI agent trying to research, analyze, write, and review its own work is like one person doing every job in a company. It works for small tasks, but for complex projects, specialization and collaboration produce far better results.
Multi-Agent Compositions assign different Level 2 patterns to different agents. A researcher agent uses ReAct with RAG to gather information. An analyst uses Tree-of-Thoughts to explore implications. A writer uses Skeleton-of-Thought for structured output. A reviewer uses Chain-of-Verification for quality control. Then these agents are coordinated through defined communication patterns — just like a human team.
Component Patterns
Each agent is powered by Level 2 compositions:
ReAct RAG Patterns Skeleton-of-Thought Meta-Prompting Reflexion Self-ConsistencyThe magic is not in any single pattern but in how specialized agents coordinate — each bringing the right capability at the right time.
Four Ways Agents Can Work Together
Sequential Handoff
Each agent completes its work and passes the result to the next. Like an assembly line — clear handoff points, predictable flow.
Supervisor Delegation
A supervisor agent decomposes the task and delegates to workers. It decides who to call, what to ask, and how to combine results.
Adversarial Review
Two advocates argue different positions across multiple rounds. A judge evaluates the debate and renders a verdict. Diverse reasoning reduces blind spots.
Parallel Voting
Multiple agents solve the same task independently using different patterns. Their answers are combined by voting or synthesis. Independence improves reliability.
Agent Roles
Researcher
Gathers info from tools and knowledge bases
Analyst
Explores implications through branching reasoning
Writer
Produces structured, organized output
Reviewer
Checks facts, catches errors, ensures quality
Supervisor
Orchestrates workers, decides delegation
Critic
Provides constructive feedback for improvement
See It in Action
A research report pipeline using the sequential topology.
Why This Works
Specialization works for the same reason it works in human teams: each agent can be optimized for its specific task. A researcher agent's prompt is tuned for thoroughness and source finding. A reviewer's prompt is tuned for skepticism and error detection. These are fundamentally different cognitive modes.
The communication topology adds a second layer of value. In a pipeline, each stage refines the previous. In a debate, adversarial pressure catches blind spots neither side would find alone. In an ensemble, independent approaches provide robustness against any single method's weaknesses.
The System
Give each agent a specialized role and the right technique for that role. Connect them through pipelines, hierarchies, debates, or ensembles. The team is greater than the sum of its parts.
When to Use This
- • Complex tasks that benefit from different types of expertise at different stages
- • Workflows that need checks and balances (writing + reviewing, coding + testing)
- • Long-running projects where modularity helps debugging and improvement
- • High-stakes tasks where adversarial review or ensemble voting improves reliability
When to Skip This
- • Simple tasks — a single agent with one pattern handles most straightforward requests
- • Latency-sensitive — multi-agent communication adds significant overhead
- • Cost-sensitive — N agents means N times the LLM calls
- • No benefit from multiple perspectives — some tasks have a single clear approach
How It Relates
Each agent in a multi-agent system can internally run a Cognitive Loop. For tool-heavy tasks, agents can delegate to JARVIS. For hard reasoning within a single agent, LATS can plug in. The Adaptive Pattern Router can choose which topology to use based on the task.
At Level 4, Federated Agent Networks extend this concept to distributed systems where agents collaborate without central control, and Collective Intelligence patterns emerge from large-scale agent interaction.