Skip to main content
hc
Submit
mcpSource-backedReview first Safety ✓ Privacy ✓

Ruflo MCP Server

MCP server and CLI for Ruflo's multi-agent Claude Code and Codex workflow harness, exposing tools for swarms, agents, memory, hooks, daemons, plugins, federation, and project automation.

by RuvNet·added 2026-06-06·
Claude CodeClaude Desktop
HarnessClaude CodeClaude Desktop
Review first — review before installing

Open the source and read safety notes before installing.

Safety notes

  • Ruflo's full CLI install can add project files, MCP configuration, hooks, helper scripts, memory state, commands, skills, and background automation.
  • The MCP surface includes tools for agent spawning, swarm coordination, memory operations, hook execution, daemon control, plugins, security scans, and federation.
  • Background daemons and hooks can run automatically based on session, task, file, or worker events; inspect generated settings before relying on them.
  • Some workflows can execute shell commands, edit project files, scan dependencies, start services, or dispatch multi-step agent tasks.
  • Federation and HTTP transport modes can expose agent coordination outside the local stdio boundary; keep them disabled unless you have reviewed network trust.
  • Use least-privilege repositories and require human review before accepting generated code, security fixes, plugin installs, or automated workflow changes.

Privacy notes

  • Ruflo memory, hooks, logs, telemetry-style metrics, worker output, prompts, code snippets, file paths, and task history may be stored under project-local state.
  • Agent memory and embeddings can preserve sensitive implementation details beyond the current chat session.
  • Federation, hosted UI, remote provider, plugin, or HTTP transport workflows may move task context beyond the local machine if enabled.
  • Security scans, dependency checks, and code analysis can reveal private package names, vulnerabilities, paths, branch names, and repository structure.
  • Review `.claude`, `.claude-flow`, helper scripts, logs, and exported memory before sharing a repository or support bundle.

Prerequisites

  • Node.js 20 or newer.
  • Claude Code or another MCP client that can launch stdio servers.
  • A repository where Ruflo-generated `.claude`, `.claude-flow`, settings, helpers, hooks, and memory files are acceptable.
  • Review of which MCP tool groups, hooks, plugins, providers, and background workers should be enabled.
  • A plan for any optional HTTP transport, federation, remote model providers, or hosted UI integrations before enabling them.

Schema details

Install type
cli
Troubleshooting
No
Source repository stats
Scope
Source repo
Collection metadata
Estimated setup
20 minutes
Difficulty
advanced
Tool listing metadata
Disclosure
MIT-licensed open source MCP server and CLI from RuvNet. Ruflo also references hosted demos, plugins, sponsorship, and broader commercial ecosystem pages; this entry covers the source-available local MCP server path.
Full copyable content
{
  "mcpServers": {
    "ruflo": {
      "command": "npx",
      "args": ["-y", "ruflo@latest", "mcp", "start"]
    }
  }
}

About this resource

Content

Ruflo MCP Server connects Claude and other MCP clients to Ruflo, a multi-agent workflow harness for Claude Code and Codex-style development environments. Its MCP mode exposes coordinated agents, swarms, memory, hooks, daemon controls, plugins, security workflows, and federation-oriented tools.

Use it when Claude needs supervised access to a broad local agent-coordination system rather than a narrow single-purpose tool. Start with the minimal MCP server configuration, then enable hooks, daemons, plugins, and federation only after reviewing their generated files and runtime behavior.

Source Review

These sources were reviewed on 2026-06-06. Prefer the live repository, README, npm metadata, license, security policy, user guide, Ruflo package metadata, CLI entrypoint, MCP bridge, and MCP package metadata for current setup and behavior details.

Features

  • Start a Ruflo MCP server for Claude Code or other stdio-compatible MCP clients.
  • Coordinate agent swarms and spawn specialized agents for coding, review, testing, docs, architecture, and security workflows.
  • Store and search persistent memory through Ruflo and AgentDB-backed workflows.
  • Use hook-based task routing, session lifecycle events, progress checks, and worker dispatch.
  • Manage plugins that add tools, hooks, workers, providers, and domain-specific workflows.
  • Run background daemon commands for audits, optimization, mapping, and worker status.
  • Configure MCP tool groups to reduce the enabled surface for specific workflows.
  • Use optional federation and HTTP transport modes when remote coordination is intentionally required.

Installation

Run the interactive initializer when you want Ruflo to set up a project:

npx ruflo@latest init wizard

For Claude Code MCP registration, use the upstream command:

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

Other MCP clients can launch the same stdio server:

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

Keep the initial install minimal until you know which hooks, daemons, plugins, memory backends, model providers, and transports your project actually needs.

Use Cases

  • Coordinate several specialized agents around a large feature, refactor, or review plan.
  • Persist useful implementation patterns and retrieve them across later Claude sessions.
  • Route tasks through review, testing, documentation, architecture, or security agent roles.
  • Run progress checks and worker status commands from a Claude Code session.
  • Add a focused plugin only after reviewing the extra MCP tools, hooks, and background behavior it enables.
  • Experiment with local-only swarm coordination before considering federation or HTTP transport.

Safety and Privacy

Ruflo is closer to an agent operating layer than a small utility server. Treat its MCP tools, generated project files, hooks, daemons, plugin installs, and memory state as changes to your development environment, not just chat context.

Run it first in a disposable or low-risk repository, inspect generated files, and limit enabled tool groups. Do not enable federation, hosted UI, HTTP transport, remote providers, or autonomous background workers until your team has reviewed what task context, code, memory, logs, and credentials could be exposed.

#agents#swarm#memory#claude-code#workflow-automation

Source citations

Signals

Loading live community signals…

More like this, weekly

A short, calm digest of reviewed Claude resources. Unsubscribe any time.