ruflo-swarm

ruflo-swarm is a Claude Code mod by ruvnet that agent teams, swarm coordination, Monitor streams, and worktree isolation — wraps 4 swarm_* + 8 agent_* MCP tools (12 total) plus 6 topologies (hierarchical / mesh / hierarchical-mesh /.

Updated 2026-10-18

What is ruflo-swarm?

ruflo-swarm is a Claude Code mod by ruvnet that agent teams, swarm coordination, Monitor streams, and worktree isolation — wraps 4 swarm_* + 8 agent_* MCP tools (12 total) plus 6 topologies (hierarchical / mesh / hierarchical-mesh /. It is listed under Cost & usage and has 76,145 GitHub stars.

TypeClaude code mod
Authorruvnet
Repositoryruvnet/ruflo
Popularity76,145 GitHub stars
CategoryCost & usage
SecurityReview before installing (64/100)
Rank#43 of 5,272 Claude Code mods; #2 in Cost & usage
LicenseMIT
Last commit2026-10-18
Stars this week+1,787
Hooks9 events
Seesevery tool call, everything, subagent spawns
Draws onPromptHint, Pane

Is ruflo-swarm safe to install?

ruflo-swarm scores 64/100 on the MCPMods security check: Review before installing. Things to review: starts processes outside the sandbox; writes files or config; watches every prompt or event. It passed 5 of 8 checks.

Mod footprints come from claude plugin validate, which lists every event a mod hooks and every API call it makes without running it.

CheckResult
No network accessPass
Starts processes outside the sandboxReview
Writes files or configReview
Doesn't read environment variablesPass
Watches every prompt or eventReview
Doesn't drive ClaudePass
Passes claude plugin validatePass
Has an open-source licensePass

API calls in its code

CallWhat it means
$.agent.list
$.clock.after
$.clock.every
$.clock.now
$.command.register
$.fs.existsreads files
$.fs.readreads files
$.fs.statreads files
$.fs.writewrites files
$.process.runruns processes
$.prompt.fillwrites the prompt box
$.session.usage
$.store.getkeeps its own state
$.store.setkeeps its own state
$.ui.closedraws interface
$.ui.invalidatedraws interface
$.ui.logdraws interface
$.ui.opendraws interface
$.ui.resolvedraws interface
$.ui.toastdraws interface

How do I install ruflo-swarm?

  1. Check your Claude Code version. Mods need Claude Code 2.1.287 or later and are on by default.
    claude --version
  2. Read its footprint. This lists every event it hooks and every API call it makes, without running it.
    git clone https://github.com/ruvnet/ruflo
    claude plugin validate ./ruflo/plugins/ruflo-swarm
  3. Install it. Run this in your terminal, then /reload-plugins in an open session.
    claude plugin marketplace add ruvnet/ruflo && claude plugin install ruflo-swarm@ruflo

Or ask Claude to set it up

Paste this into Claude Code and it installs ruflo-swarm for you:

Install the "ruflo-swarm" Claude Code mod from https://github.com/ruvnet/ruflo/tree/main/plugins/ruflo-swarm. Run claude plugin validate on it first and show me what it can access before you enable it.

More from ruvnet/ruflo

This repository has at least 9 Claude Code mods on MCPMods.

  • ruflo-adr: ADR lifecycle management — create, index, reconcile, supersede, check compliance, and link Architecture Decision Records (Low risk)
  • ruflo-agent: Agent runtimes for ruflo — local WASM-sandboxed agents (rvagent: 10 wasm_agent_*/wasm_gallery_* MCP tools, built on @ruv (Low risk)
  • ruflo-agentdb: Substrate plugin for Ruflo memory: AgentDB controller bridge (15 agentdb_* MCP tools), RuVector ONNX embeddings (10 embe (High risk)
  • ruflo-agntcy: AGNTCY/Outshift Internet-of-Cognition runtime integration for Ruflo — SLIM secure transport for cross-host agent coordin (Low risk)
  • ruflo-ai-team: Give Claude a tenant-isolated AI team that divides work, coordinates progress, recalls approved context, and returns evi (Low risk)
  • ruflo-aidefence: AI safety scanning, PII detection, prompt injection defense, and adaptive threat learning (Low risk)
  • ruflo-arena: Competitive ruliology for Ruflo swarms — arenas, tournaments, and adaptive co-evolution of program strategies (ADR-147/1 (Low risk)
  • ruflo-autopilot: Autonomous /loop-driven task completion with learning, prediction, and progress tracking — wraps 10 autopilot_* MCP tool (Low risk)