Flow Nexus Swarm

Deploy AI agent swarms with cloud-based workflow automation

✨ The solution you've been looking for

Verified

Tested and verified by our team

11981 Stars

Cloud-based AI swarm deployment and event-driven workflow automation with Flow Nexus platform

swarm workflow cloud agents automation orchestration message-queue devops

Repository

See It In Action

Interactive preview & real-world examples

Live Demo

AI Conversation Simulator

See how users interact with this skill

User Prompt

Create a swarm to build an e-commerce platform with Node.js, React, and PostgreSQL. I need backend, frontend, database, and QA specialists working together.

Skill Processing

Analyzing request...

Agent Response

A coordinated team of AI agents handling requirements analysis, database design, API development, UI creation, integration, and testing in parallel workflows

Quick Start (3 Steps)

Get up and running in minutes

Install

claude-code skill install flow-nexus-swarm

claude-code skill install flow-nexus-swarm

Config

First Trigger

@flow-nexus-swarm help

Commands

Command	Description	Required Args
@flow-nexus-swarm full-stack-development-team	Deploy a hierarchical swarm with specialized agents for complete web development projects	None
@flow-nexus-swarm ci/cd-pipeline-automation	Set up automated testing, building, and deployment workflows with message queue processing	None
@flow-nexus-swarm research-&-analysis-swarm	Deploy a mesh topology for collaborative research with parallel information gathering	None

Typical Use Cases

Full-Stack Development Team

Deploy a hierarchical swarm with specialized agents for complete web development projects

CI/CD Pipeline Automation

Set up automated testing, building, and deployment workflows with message queue processing

Research & Analysis Swarm

Deploy a mesh topology for collaborative research with parallel information gathering

Overview

Flow Nexus Swarm & Workflow Orchestration

Deploy and manage cloud-based AI agent swarms with event-driven workflow automation, message queue processing, and intelligent agent coordination.

📋 Table of Contents

Overview

Flow Nexus provides cloud-based orchestration for AI agent swarms with:

Multi-topology Support: Hierarchical, mesh, ring, and star architectures
Event-driven Workflows: Message queue processing with async execution
Template Library: Pre-built swarm configurations for common use cases
Intelligent Agent Assignment: Vector similarity matching for optimal agent selection
Real-time Monitoring: Comprehensive metrics and audit trails
Scalable Infrastructure: Cloud-based execution with auto-scaling

Swarm Management

Initialize Swarm

Create a new swarm with specified topology and configuration:

1mcp__flow-nexus__swarm_init({
2  topology: "hierarchical", // Options: mesh, ring, star, hierarchical
3  maxAgents: 8,
4  strategy: "balanced" // Options: balanced, specialized, adaptive
5})

Topology Guide:

Hierarchical: Tree structure with coordinator nodes (best for complex projects)
Mesh: Peer-to-peer collaboration (best for research and analysis)
Ring: Circular coordination (best for sequential workflows)
Star: Centralized hub (best for simple delegation)

Strategy Guide:

Balanced: Equal distribution of workload across agents
Specialized: Agents focus on specific expertise areas
Adaptive: Dynamic adjustment based on task complexity

Spawn Agents

Add specialized agents to the swarm:

1mcp__flow-nexus__agent_spawn({
2  type: "researcher", // Options: researcher, coder, analyst, optimizer, coordinator
3  name: "Lead Researcher",
4  capabilities: ["web_search", "analysis", "summarization"]
5})

Agent Types:

Researcher: Information gathering, web search, analysis
Coder: Code generation, refactoring, implementation
Analyst: Data analysis, pattern recognition, insights
Optimizer: Performance tuning, resource optimization
Coordinator: Task delegation, progress tracking, integration

Orchestrate Tasks

Distribute tasks across the swarm:

1mcp__flow-nexus__task_orchestrate({
2  task: "Build a REST API with authentication and database integration",
3  strategy: "parallel", // Options: parallel, sequential, adaptive
4  maxAgents: 5,
5  priority: "high" // Options: low, medium, high, critical
6})

Execution Strategies:

Parallel: Maximum concurrency for independent subtasks
Sequential: Step-by-step execution with dependencies
Adaptive: AI-powered strategy selection based on task analysis

Monitor & Scale Swarms

 1// Get detailed swarm status
 2mcp__flow-nexus__swarm_status({
 3  swarm_id: "optional-id" // Uses active swarm if not provided
 4})
 5
 6// List all active swarms
 7mcp__flow-nexus__swarm_list({
 8  status: "active" // Options: active, destroyed, all
 9})
10
11// Scale swarm up or down
12mcp__flow-nexus__swarm_scale({
13  target_agents: 10,
14  swarm_id: "optional-id"
15})
16
17// Gracefully destroy swarm
18mcp__flow-nexus__swarm_destroy({
19  swarm_id: "optional-id"
20})

Workflow Automation

Create Workflow

Define event-driven workflows with message queue processing:

 1mcp__flow-nexus__workflow_create({
 2  name: "CI/CD Pipeline",
 3  description: "Automated testing, building, and deployment",
 4  steps: [
 5    {
 6      id: "test",
 7      action: "run_tests",
 8      agent: "tester",
 9      parallel: true
10    },
11    {
12      id: "build",
13      action: "build_app",
14      agent: "builder",
15      depends_on: ["test"]
16    },
17    {
18      id: "deploy",
19      action: "deploy_prod",
20      agent: "deployer",
21      depends_on: ["build"]
22    }
23  ],
24  triggers: ["push_to_main", "manual_trigger"],
25  metadata: {
26    priority: 10,
27    retry_policy: "exponential_backoff"
28  }
29})

Workflow Features:

Dependency Management: Define step dependencies with depends_on
Parallel Execution: Set parallel: true for concurrent steps
Event Triggers: GitHub events, schedules, manual triggers
Retry Policies: Automatic retry on transient failures
Priority Queuing: High-priority workflows execute first

Execute Workflow

Run workflows synchronously or asynchronously:

1mcp__flow-nexus__workflow_execute({
2  workflow_id: "workflow_id",
3  input_data: {
4    branch: "main",
5    commit: "abc123",
6    environment: "production"
7  },
8  async: true // Queue-based execution for long-running workflows
9})

Execution Modes:

Sync (async: false): Immediate execution, wait for completion
Async (async: true): Message queue processing, non-blocking

Monitor Workflows

 1// Get workflow status and metrics
 2mcp__flow-nexus__workflow_status({
 3  workflow_id: "id",
 4  execution_id: "specific-run-id", // Optional
 5  include_metrics: true
 6})
 7
 8// List workflows with filters
 9mcp__flow-nexus__workflow_list({
10  status: "running", // Options: running, completed, failed, pending
11  limit: 10,
12  offset: 0
13})
14
15// Get complete audit trail
16mcp__flow-nexus__workflow_audit_trail({
17  workflow_id: "id",
18  limit: 50,
19  start_time: "2025-01-01T00:00:00Z"
20})

Agent Assignment

Intelligently assign agents to workflow tasks:

1mcp__flow-nexus__workflow_agent_assign({
2  task_id: "task_id",
3  agent_type: "coder", // Preferred agent type
4  use_vector_similarity: true // AI-powered capability matching
5})

Vector Similarity Matching:

Analyzes task requirements and agent capabilities
Finds optimal agent based on past performance
Considers workload and availability

Queue Management

Monitor and manage message queues:

1mcp__flow-nexus__workflow_queue_status({
2  queue_name: "optional-specific-queue",
3  include_messages: true // Show pending messages
4})

Agent Orchestration

Full-Stack Development Pattern

 1// 1. Initialize swarm with hierarchical topology
 2mcp__flow-nexus__swarm_init({
 3  topology: "hierarchical",
 4  maxAgents: 8,
 5  strategy: "specialized"
 6})
 7
 8// 2. Spawn specialized agents
 9mcp__flow-nexus__agent_spawn({ type: "coordinator", name: "Project Manager" })
10mcp__flow-nexus__agent_spawn({ type: "coder", name: "Backend Developer" })
11mcp__flow-nexus__agent_spawn({ type: "coder", name: "Frontend Developer" })
12mcp__flow-nexus__agent_spawn({ type: "coder", name: "Database Architect" })
13mcp__flow-nexus__agent_spawn({ type: "analyst", name: "QA Engineer" })
14
15// 3. Create development workflow
16mcp__flow-nexus__workflow_create({
17  name: "Full-Stack Development",
18  steps: [
19    { id: "requirements", action: "analyze_requirements", agent: "coordinator" },
20    { id: "db_design", action: "design_schema", agent: "Database Architect" },
21    { id: "backend", action: "build_api", agent: "Backend Developer", depends_on: ["db_design"] },
22    { id: "frontend", action: "build_ui", agent: "Frontend Developer", depends_on: ["requirements"] },
23    { id: "integration", action: "integrate", agent: "Backend Developer", depends_on: ["backend", "frontend"] },
24    { id: "testing", action: "qa_testing", agent: "QA Engineer", depends_on: ["integration"] }
25  ]
26})
27
28// 4. Execute workflow
29mcp__flow-nexus__workflow_execute({
30  workflow_id: "workflow_id",
31  input_data: {
32    project: "E-commerce Platform",
33    tech_stack: ["Node.js", "React", "PostgreSQL"]
34  }
35})

Research & Analysis Pattern

 1// 1. Initialize mesh topology for collaborative research
 2mcp__flow-nexus__swarm_init({
 3  topology: "mesh",
 4  maxAgents: 5,
 5  strategy: "balanced"
 6})
 7
 8// 2. Spawn research agents
 9mcp__flow-nexus__agent_spawn({ type: "researcher", name: "Primary Researcher" })
10mcp__flow-nexus__agent_spawn({ type: "researcher", name: "Secondary Researcher" })
11mcp__flow-nexus__agent_spawn({ type: "analyst", name: "Data Analyst" })
12mcp__flow-nexus__agent_spawn({ type: "analyst", name: "Insights Analyst" })
13
14// 3. Orchestrate research task
15mcp__flow-nexus__task_orchestrate({
16  task: "Research machine learning trends for 2025 and analyze market opportunities",
17  strategy: "parallel",
18  maxAgents: 4,
19  priority: "high"
20})

CI/CD Pipeline Pattern

 1mcp__flow-nexus__workflow_create({
 2  name: "Deployment Pipeline",
 3  description: "Automated testing, building, and multi-environment deployment",
 4  steps: [
 5    { id: "lint", action: "lint_code", agent: "code_quality", parallel: true },
 6    { id: "unit_test", action: "unit_tests", agent: "test_runner", parallel: true },
 7    { id: "integration_test", action: "integration_tests", agent: "test_runner", parallel: true },
 8    { id: "build", action: "build_artifacts", agent: "builder", depends_on: ["lint", "unit_test", "integration_test"] },
 9    { id: "security_scan", action: "security_scan", agent: "security", depends_on: ["build"] },
10    { id: "deploy_staging", action: "deploy", agent: "deployer", depends_on: ["security_scan"] },
11    { id: "smoke_test", action: "smoke_tests", agent: "test_runner", depends_on: ["deploy_staging"] },
12    { id: "deploy_prod", action: "deploy", agent: "deployer", depends_on: ["smoke_test"] }
13  ],
14  triggers: ["github_push", "github_pr_merged"],
15  metadata: {
16    priority: 10,
17    auto_rollback: true
18  }
19})

Data Processing Pipeline Pattern

 1mcp__flow-nexus__workflow_create({
 2  name: "ETL Pipeline",
 3  description: "Extract, Transform, Load data processing",
 4  steps: [
 5    { id: "extract", action: "extract_data", agent: "data_extractor" },
 6    { id: "validate_raw", action: "validate_data", agent: "validator", depends_on: ["extract"] },
 7    { id: "transform", action: "transform_data", agent: "transformer", depends_on: ["validate_raw"] },
 8    { id: "enrich", action: "enrich_data", agent: "enricher", depends_on: ["transform"] },
 9    { id: "load", action: "load_data", agent: "loader", depends_on: ["enrich"] },
10    { id: "validate_final", action: "validate_data", agent: "validator", depends_on: ["load"] }
11  ],
12  triggers: ["schedule:0 2 * * *"], // Daily at 2 AM
13  metadata: {
14    retry_policy: "exponential_backoff",
15    max_retries: 3
16  }
17})

Templates & Patterns

Use Pre-built Templates

 1// Create swarm from template
 2mcp__flow-nexus__swarm_create_from_template({
 3  template_name: "full-stack-dev",
 4  overrides: {
 5    maxAgents: 6,
 6    strategy: "specialized"
 7  }
 8})
 9
10// List available templates
11mcp__flow-nexus__swarm_templates_list({
12  category: "quickstart", // Options: quickstart, specialized, enterprise, custom, all
13  includeStore: true
14})

Available Template Categories:

Quickstart Templates:

full-stack-dev: Complete web development swarm
research-team: Research and analysis swarm
code-review: Automated code review swarm
data-pipeline: ETL and data processing

Specialized Templates:

ml-development: Machine learning project swarm
mobile-dev: Mobile app development
devops-automation: Infrastructure and deployment
security-audit: Security analysis and testing

Enterprise Templates:

enterprise-migration: Large-scale system migration
multi-repo-sync: Multi-repository coordination
compliance-review: Regulatory compliance workflows
incident-response: Automated incident management

Custom Template Creation

Save successful swarm configurations as reusable templates for future projects.

Advanced Features

Real-time Monitoring

 1// Subscribe to execution streams
 2mcp__flow-nexus__execution_stream_subscribe({
 3  stream_type: "claude-flow-swarm",
 4  deployment_id: "deployment_id"
 5})
 6
 7// Get execution status
 8mcp__flow-nexus__execution_stream_status({
 9  stream_id: "stream_id"
10})
11
12// List files created during execution
13mcp__flow-nexus__execution_files_list({
14  stream_id: "stream_id",
15  created_by: "claude-flow"
16})

Swarm Metrics & Analytics

 1// Get swarm performance metrics
 2mcp__flow-nexus__swarm_status({
 3  swarm_id: "id"
 4})
 5
 6// Analyze workflow efficiency
 7mcp__flow-nexus__workflow_status({
 8  workflow_id: "id",
 9  include_metrics: true
10})

Multi-Swarm Coordination

Coordinate multiple swarms for complex, multi-phase projects:

 1// Phase 1: Research swarm
 2const researchSwarm = await mcp__flow-nexus__swarm_init({
 3  topology: "mesh",
 4  maxAgents: 4
 5})
 6
 7// Phase 2: Development swarm
 8const devSwarm = await mcp__flow-nexus__swarm_init({
 9  topology: "hierarchical",
10  maxAgents: 8
11})
12
13// Phase 3: Testing swarm
14const testSwarm = await mcp__flow-nexus__swarm_init({
15  topology: "star",
16  maxAgents: 5
17})

Best Practices

1. Choose the Right Topology

 1// Simple projects: Star
 2mcp__flow-nexus__swarm_init({ topology: "star", maxAgents: 3 })
 3
 4// Collaborative work: Mesh
 5mcp__flow-nexus__swarm_init({ topology: "mesh", maxAgents: 5 })
 6
 7// Complex projects: Hierarchical
 8mcp__flow-nexus__swarm_init({ topology: "hierarchical", maxAgents: 10 })
 9
10// Sequential workflows: Ring
11mcp__flow-nexus__swarm_init({ topology: "ring", maxAgents: 4 })

2. Optimize Agent Assignment

1// Use vector similarity for optimal matching
2mcp__flow-nexus__workflow_agent_assign({
3  task_id: "complex-task",
4  use_vector_similarity: true
5})

3. Implement Proper Error Handling

 1mcp__flow-nexus__workflow_create({
 2  name: "Resilient Workflow",
 3  steps: [...],
 4  metadata: {
 5    retry_policy: "exponential_backoff",
 6    max_retries: 3,
 7    timeout: 300000, // 5 minutes
 8    on_failure: "notify_and_rollback"
 9  }
10})

4. Monitor and Scale

1// Regular monitoring
2const status = await mcp__flow-nexus__swarm_status()
3
4// Scale based on workload
5if (status.workload > 0.8) {
6  await mcp__flow-nexus__swarm_scale({ target_agents: status.agents + 2 })
7}

5. Use Async Execution for Long-Running Workflows

1// Long-running workflows should use message queues
2mcp__flow-nexus__workflow_execute({
3  workflow_id: "data-pipeline",
4  async: true // Non-blocking execution
5})
6
7// Monitor progress
8mcp__flow-nexus__workflow_queue_status({ include_messages: true })

6. Clean Up Resources

1// Destroy swarm when complete
2mcp__flow-nexus__swarm_destroy({ swarm_id: "id" })

7. Leverage Templates

1// Use proven templates instead of building from scratch
2mcp__flow-nexus__swarm_create_from_template({
3  template_name: "code-review",
4  overrides: { maxAgents: 4 }
5})

Integration with Claude Flow

Flow Nexus swarms integrate seamlessly with Claude Flow hooks:

1# Pre-task coordination setup
2npx claude-flow@alpha hooks pre-task --description "Initialize swarm"
3
4# Post-task metrics export
5npx claude-flow@alpha hooks post-task --task-id "swarm-execution"

Common Use Cases

1. Multi-Repo Development

Coordinate development across multiple repositories
Synchronized testing and deployment
Cross-repo dependency management

2. Research Projects

Distributed information gathering
Parallel analysis of different data sources
Collaborative synthesis and reporting

3. DevOps Automation

Infrastructure as Code deployment
Multi-environment testing
Automated rollback and recovery

4. Code Quality Workflows

Automated code review
Security scanning
Performance benchmarking

5. Data Processing

Large-scale ETL pipelines
Real-time data transformation
Data validation and quality checks

Authentication & Setup

 1# Install Flow Nexus
 2npm install -g flow-nexus@latest
 3
 4# Register account
 5npx flow-nexus@latest register
 6
 7# Login
 8npx flow-nexus@latest login
 9
10# Add MCP server to Claude Code
11claude mcp add flow-nexus npx flow-nexus@latest mcp start

Support & Resources

Platform: https://flow-nexus.ruv.io
Documentation: https://github.com/ruvnet/flow-nexus
Issues: https://github.com/ruvnet/flow-nexus/issues

Remember: Flow Nexus provides cloud-based orchestration infrastructure. For local execution and coordination, use the core claude-flow MCP server alongside Flow Nexus for maximum flexibility.

What Users Are Saying

Real feedback from the community

Environment Matrix

Dependencies

flow-nexus MCP server

Active Flow Nexus account (register at flow-nexus.ruv.io)

Node.js (for Flow Nexus CLI)

Framework Support

Claude Flow ✓ (recommended) GitHub Events ✓ Message Queue Systems ✓

Context Window

Token Usage ~5K-15K tokens for complex swarm configurations and workflows

Security & Privacy

Information

Author: ruvnet
Updated: 2026-01-30
Category: productivity-tools

Flow Nexus Swarm

See It In Action

AI Conversation Simulator

Quick Start (3 Steps)

Install

Config

First Trigger

Commands

Typical Use Cases

Full-Stack Development Team

CI/CD Pipeline Automation

Research & Analysis Swarm

Overview

Flow Nexus Swarm & Workflow Orchestration

📋 Table of Contents

Overview

Swarm Management

Initialize Swarm

Spawn Agents

Orchestrate Tasks

Monitor & Scale Swarms

Workflow Automation

Create Workflow

Execute Workflow

Monitor Workflows

Agent Assignment

Queue Management

Agent Orchestration

Full-Stack Development Pattern

Research & Analysis Pattern

CI/CD Pipeline Pattern

Data Processing Pipeline Pattern

Templates & Patterns

Use Pre-built Templates

Custom Template Creation

Advanced Features

Real-time Monitoring

Swarm Metrics & Analytics

Multi-Swarm Coordination

Best Practices

1. Choose the Right Topology

2. Optimize Agent Assignment

3. Implement Proper Error Handling

4. Monitor and Scale

5. Use Async Execution for Long-Running Workflows

6. Clean Up Resources

7. Leverage Templates

Integration with Claude Flow

Common Use Cases

1. Multi-Repo Development

2. Research Projects

3. DevOps Automation

4. Code Quality Workflows

5. Data Processing

Authentication & Setup

Support & Resources

What Users Are Saying

Environment Matrix

Dependencies

Framework Support

Context Window

Security & Privacy

Information

Related Skills

Flow Nexus Swarm

Deployment Pipeline Design

Deployment Pipeline Design