Agent Skills: swarm-orchestration

Multi-agent swarm formation and coordinated execution with topology-aware agent deployment, consensus protocols, and anti-drift enforcement.

UncategorizedID: a5c-ai/babysitter/swarm-orchestration

Install this agent skill to your local

pnpm dlx add-skill https://github.com/a5c-ai/babysitter/tree/HEAD/library/methodologies/ruflo/skills/swarm-orchestration

Skill Files

Browse the full folder contents for swarm-orchestration.

Download Skill

Loading file tree…

library/methodologies/ruflo/skills/swarm-orchestration/SKILL.md

Skill Metadata

Name
swarm-orchestration
Description
Multi-agent swarm formation and coordinated execution with topology-aware agent deployment, consensus protocols, and anti-drift enforcement.
  • Tasks needing coordinated parallel execution
  • When consensus among agents is required for quality
  • Projects requiring anti-drift enforcement during execution

Process

  1. Topology Selection - Analyze task and agent pool to select optimal topology
  2. Agent Assignment - Assign Queen (Strategic/Tactical/Adaptive) and Worker roles
  3. Consensus Init - Initialize Raft/Byzantine/Gossip/CRDT protocol
  4. Parallel Execution - Distribute subtasks with shared memory
  5. Anti-Drift Checkpoints - Validate alignment every N subtasks
  6. Consensus Voting - Weighted voting (Queen=3x) for final decision

Topologies

  • Mesh: All-to-all communication, best for small swarms (<8 agents)
  • Hierarchical: Queen coordinates workers, best for large/structured tasks
  • Ring: Sequential handoff, best for pipeline/transformation tasks
  • Star: Central coordinator fan-out, best for independent subtasks

Agents Used

  • agents/strategic-queen/ - Long-term planning swarms
  • agents/tactical-queen/ - Execution coordination swarms
  • agents/adaptive-queen/ - Real-time optimization swarms
  • agents/swarm-coordinator/ - Topology management

Tool Use

Invoke via babysitter process: methodologies/ruflo/ruflo-swarm-coordination