Research: Multi-Agent Coordination Patterns from Industry

[wonderwomancode]

Overview

Document and apply industry research on multi-agent coordination patterns.

Sources to Study

Anthropic

• How we built our multi-agent research system
• Claude Code: Best practices for agentic coding

ByteDance

• MegaAgent - O(log n) parallel cooperation
• DeerFlow - Modular multi-agent

AllenAI

• Asta - Agent evaluation
• ADAS (meta-agents designing new agents)
• DynaSaur (dynamic action programs)

Community

• lizTheDeveloper/ai-agent-messaging - NATS + MCP patterns
• livecode_agent_swarm_demo - LangGraph swarm

Patterns to Document

• Orchestrator-Worker (Anthropic)
• Effort Scaling Rules (Anthropic)
• Dynamic Task Decomposition (MegaAgent)
• Hierarchical Async Execution (DeerFlow)
• Agent Relationship Dynamics (lizTheDeveloper)
• Prompt Injection Detection (livecode swarm)

Acceptance Criteria

• Read and summarize each source
• Identify patterns applicable to our swarm
• Update AGENTS.md with new patterns
• Create implementation recommendations
• Track which patterns we've adopted

Assigned Agent

@senku - Architecture research
@lain - Security pattern review