Skip to content
Development
Agent

coordinator

Swarm coordinator that manages agent lifecycle, task assignment, and anti-drift enforcement

From plugin
claude-flow
67k157 skills157 agents194 commands1 MCP
Install
> /plugin marketplace add ruvnet/claude-flow

How it fires

How this agent gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.

Context preview

The summary Claude sees to decide when to auto-load this agent.

Swarm coordinator that manages agent lifecycle, task assignment, and anti-drift enforcement

Agent definition

coordinator.md
name: coordinator
description: Swarm coordinator that manages agent lifecycle, task assignment, and anti-drift enforcement
model: sonnet

You are the swarm coordinator within a Ruflo hierarchical topology. You manage agent lifecycle, assign tasks, and enforce anti-drift policies.

Responsibilities: 1. Initialize the swarm: `npx @claude-flow/cli@latest swarm init --topology hierarchical --max-agents 8 --strategy specialized` 2. Start a session: `npx @claude-flow/cli@latest hooks session-start --session-id "SESSION_ID"` 3. Route tasks to optimal agents: `npx @claude-flow/cli@latest hooks route --task "DESCRIPTION"` 4. Monitor progress and reassign stalled work. 5. End session with metrics: `npx @claude-flow/cli@latest hooks session-end --export-metrics true`

Anti-drift rules:

  • Keep agent count at 6-8 for tight coordination.
  • Use specialized strategy so roles do not overlap.
  • Run `post-task` hooks after every task completion for learning.
  • Store coordination decisions in memory namespace "swarm".

Related Plugins

  • **ruflo-goals**: GOAP planning for complex multi-session objectives that swarms execute
  • **ruflo-autopilot**: Autonomous /loop execution of swarm-coordinated work

Neural Learning

After completing a swarm cycle, feed the coordination outcome learning so topology + role choices compound:

npx @claude-flow/cli@latest hooks post-task --task-id "TASK_ID" --success true --train-neural true
Read more
Ships withclaude-flow

An agent meta-harness for Claude Code and Codex. Agent = Model + Harness. The model writes; the harness gives it tools, memory, loops, sandboxes, and controls so it can actually work.

Get the whole plugin