mcpField
S
swarm-init

swarm-init

github.com/ruvnetAgent workflows80.0k747 installsUpdated today
Repository

About swarm-init

swarm-init is a skill for initializing hierarchical swarms for coordinated multi-agent work. It describes setup through MCP or the CLI, then guidance for spawning and coordinating named agents in parallel.

Key features

  1. 01Initialize hierarchical swarms
  2. 02Configure agent limits and strategy
  3. 03Spawn named agents in parallel
  4. 04Coordinate agents with SendMessage
  5. 05Use worktrees for parallel Git work

Use cases

  1. 01Set up a coordinated multi-agent project
  2. 02Scale a team to 10 or more agents
  3. 03Run parallel agent tasks with Git isolation

MCP support

The skill provides an MCP swarm_init call for creating a hierarchical swarm, alongside a CLI alternative.

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/ruvnet/ruflo --skill swarm-init

After installing, type/swarm-initin Claude Code to confirm it loaded.

Details

From the project README

Initialize a hierarchical swarm for coordinated multi-agent work.

Via MCP: mcp__plugin_ruflo-core_ruflo__swarm_init({ topology: "hierarchical", maxAgents: 8, strategy: "specialized" })

Or via CLI:

npx @claude-flow/cli@latest swarm init --topology hierarchical --max-agents 8 --strategy specialized

Then spawn named agents in ONE message via Claude Code's Task tool with name: (for SendMessage addressability) and run_in_background: true (for parallel execution). Use EnterWorktree per agent for git-safe parallel work, and SendMessage for inter-agent coordination.

For larger teams (10+), use hierarchical-mesh topology:

npx @claude-flow/cli@latest swarm init --topology hierarchical-mesh --max-agents 15 --strategy specialized

Comments

Markdown supported