Ruflo

v3.7.0-alpha.5Coding Agentsstable

🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat

agentic-aiagentic-frameworkagentic-ragagentic-workflowagents
Share:
53,981
Stars
0
Downloads
0
Weekly
0/5

What is Ruflo?

Ruflo is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to 🌊 the leading agent orchestration platform for claude. deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational ai systems. features enterprise-grade architect...

🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat

This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • 🌊 The leading agent orchestration platform for Claude. Deplo

Use Cases

Deploy multi-agent AI swarms that coordinate complex workflows.
Build conversational AI systems with self-learning intelligence and RAG integration.
ruvnet

Maintainer

LicenseMIT
Languagetypescript
Versionv3.7.0-alpha.5
UpdatedMay 22, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y @claude-flow/cli-core

Manual Installation

npx -y @claude-flow/cli-core

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Ruflo

Ruflo is a multi-agent orchestration platform built on top of Claude Code that coordinates networks of 100+ specialized AI agents across machines and trust boundaries. It adds persistent vector memory with HNSW indexing, self-learning via SONA neural patterns, federated agent communication with zero-trust authentication, and ~210 MCP tools organized into five server groups. Engineers and teams use Ruflo to move beyond single-agent Claude Code sessions into collaborative swarm architectures that retain knowledge across sessions and automatically route tasks to the most suitable specialist agent.

Prerequisites

  • Node.js 18 or later and npx available in your PATH
  • Claude Code installed and authenticated with a valid Anthropic API key
  • Git installed (Ruflo stores configuration in .claude/ and .claude-flow/ directories)
  • Optional: access to goal.ruv.io or flo.ruv.io for the web UI and goal planner features
1

Initialize Ruflo in your project

Run the Ruflo init wizard to set up the .claude/ and .claude-flow/ configuration directories, select your LLM providers, and configure memory namespaces and security profiles.

npx ruflo@latest init wizard
2

Register Ruflo as an MCP server in Claude Code

Add Ruflo as an MCP server so Claude Code can access all 210+ MCP tools it provides, including agent spawning, memory operations, and federation commands.

claude mcp add ruflo -- npx ruflo@latest mcp start
3

Start background workers

Ruflo's 12 background workers handle automated audits, test generation, optimization, and agent health checks. Start them after initialization.

npx ruflo@latest workers start
4

Use Claude Code normally — Ruflo operates in the background

Once registered, Ruflo's hooks system automatically intercepts Claude Code tasks, routes them to specialist agents, stores results in vector memory, and learns from patterns. No explicit commands are needed for basic use.

5

Use CLI commands for direct swarm and memory operations

Ruflo exposes 60+ CLI commands for explicit control over agents and memory when needed.

# Spawn a swarm for a specific task
npx ruflo@latest swarm_init --task "audit this codebase for security issues"

# Store something in vector memory
npx ruflo@latest memory_store --key "arch-decisions" --value "We use PostgreSQL for persistence"

# Search memory
npx ruflo@latest memory_search --query "database architecture"

Ruflo Examples

Client configuration

MCP server entry for Claude Code's configuration to register Ruflo's full tool set.

{
  "mcpServers": {
    "ruflo": {
      "command": "npx",
      "args": ["ruflo@latest", "mcp", "start"]
    }
  }
}

Prompts to try

Example prompts that leverage Ruflo's agent orchestration, memory, and swarm capabilities within Claude Code.

- "Spawn a security audit swarm to review all TypeScript files in src/"
- "Search my project memory for previous architecture decisions about the database"
- "Coordinate three agents to write, test, and document the new auth module in parallel"
- "Store this design decision in long-term memory: we use JWT for stateless auth"
- "Show me the status of all running agents and their tasks"

Troubleshooting Ruflo

npx ruflo@latest fails with 'package not found' or network error

Check your npm registry with 'npm view ruflo'. If the package is private or not yet published under that exact name, use the curl installer instead: 'curl -fsSL https://cdn.jsdelivr.net/gh/ruvnet/ruflo@main/scripts/install.sh | bash'.

MCP tools are not appearing in Claude Code after 'claude mcp add ruflo'

Ensure Claude Code is fully restarted after adding the MCP server. Run 'claude mcp list' to verify ruflo is registered. If the server crashes on start, check that Node.js 18+ is installed and run 'npx ruflo@latest mcp start' directly to see any error output.

Background workers stop after terminal session ends

Run workers with a process manager: 'npx ruflo@latest workers start --daemon' or configure a launchd/systemd service to keep workers running persistently across sessions.

Frequently Asked Questions about Ruflo

What is Ruflo?

Ruflo is a Model Context Protocol (MCP) server that 🌊 the leading agent orchestration platform for claude. deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational ai systems. features enterprise-grade architecture, self-learning swarm intelligence, rag integrat It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Ruflo?

Install via npm with the command: npx -y @claude-flow/cli-core. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with Ruflo?

Ruflo works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Ruflo free to use?

Yes, Ruflo is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Coding Agents MCP Servers

Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "ruflo": { "command": "npx", "args": ["-y", "@claude-flow/cli-core"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Ruflo?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides