Hive Mind — многоагентная координация с коллективным интеллектом

ai-tooling

hive-mind-advanced is a Claude Code skill that implements the advanced Hive Mind collective intelligence system for multi-agent coordination inside Claude Flow. The architecture is built on a queen hierarchy with three types — strategic, tactical, and adaptive — directing eight specialized worker roles: researcher, coder, analyst, tester, architect, reviewer, optimizer, and documenter. Decisions are reached through three consensus algorithms: simple majority, weighted voting where a queen's vote counts three times, and Byzantine Fault Tolerance requiring a 2/3 supermajority. Collective memory persists in SQLite with WAL mode and an LRU cache, supporting eight record types with distinct TTLs and pattern-based search with confidence filtering. Teams tackling complex orchestration challenges — from microservices architecture to full-stack development cycles — can manage everything through the npx claude-flow hive-mind CLI with session checkpointing, auto-scaling up to 12 workers, and exportable session logs.