Skip to content

mangel99/HIVEMIND

BranchesTags

Repository files navigation


🐝 HIVEMIND

Multi-Agent AI Orchestration System


License: MIT Python Agents Teams


One command. 24 AI agents. Infinite possibilities.


Quick Start • The Hive • Commands • Examples • Install



🧠 What is HIVEMIND?

HIVEMIND coordinates 24 specialized AI agents across 4 teams to accomplish complex technical tasks.

Instead of one AI doing everything, HIVEMIND routes your request to the right specialists—architects, security experts, infrastructure engineers, and QA professionals—who work together to deliver production-grade results.


                         YOUR REQUEST
                              │
                              ▼
                    ┌─────────────────┐
                    │   HEAD_CODEX    │
                    │  ─────────────  │
                    │   Orchestrator  │
                    └────────┬────────┘
                             │
           ┌─────────────────┼─────────────────┐
           │                 │                 │
           ▼                 ▼                 ▼
     ┌───────────┐    ┌───────────┐    ┌───────────┐
     │    DEV    │    │    SEC    │    │    INF    │
     │ 6 Agents  │    │ 6 Agents  │    │ 6 Agents  │
     └─────┬─────┘    └─────┬─────┘    └─────┬─────┘
           │                │                │
           └────────────────┼────────────────┘
                            │
                            ▼
                    ┌───────────────┐
                    │    QA TEAM    │
                    │   6 Agents    │
                    └───────┬───────┘
                            │
                            ▼
                    ┌───────────────┐
                    │ QUALITY GATES │
                    │ G1→G2→G3→G4→G5│
                    └───────┬───────┘
                            │
                            ▼
                    ┌───────────────┐
                    │    REPORT     │
                    └───────────────┘


⚡ Quick Start

git clone https://github.com/Nerds489/HIVEMIND.git
cd HIVEMIND
./install.sh
hivemind

That's it. You're in.



🐝 The Hive

💻 Development Team — DEV

Build it right.

Agent Role Specialty
DEV-001 Architect System design, patterns, APIs, microservices
DEV-002 Backend Developer APIs, servers, databases, Python, Node.js
DEV-003 Frontend Developer UI/UX, React, Vue, Angular, TypeScript
DEV-004 Code Reviewer Code quality, pull requests, best practices
DEV-005 Technical Writer Documentation, guides, API docs
DEV-006 DevOps Liaison CI/CD, pipelines, GitHub Actions

🔐 Security Team — SEC

Break it before they do.

Agent Role Specialty
SEC-001 Security Architect Threat modeling, security design, risk
SEC-002 Penetration Tester Vulnerability assessment, OWASP, exploits
SEC-003 Malware Analyst Reverse engineering, threat analysis, IOCs
SEC-004 Wireless Security WiFi, Bluetooth, RF, IoT security
SEC-005 Compliance Auditor SOC2, GDPR, PCI-DSS, NIST
SEC-006 Incident Responder Breach response, forensics, containment

☁️ Infrastructure Team — INF

Run it at scale.

Agent Role Specialty
INF-001 Infrastructure Architect Cloud design, AWS, GCP, Azure
INF-002 Systems Administrator Linux, Windows, configuration
INF-003 Network Engineer Firewalls, DNS, routing, VPCs
INF-004 Database Administrator PostgreSQL, MySQL, MongoDB
INF-005 Site Reliability Engineer Kubernetes, monitoring, SLOs
INF-006 Automation Engineer Terraform, Ansible, IaC

✅ Quality Assurance Team — QA

Prove it works.

Agent Role Specialty
QA-001 QA Architect Test strategy, coverage, quality planning
QA-002 Test Automation Selenium, Cypress, pytest, frameworks
QA-003 Performance Tester Load testing, JMeter, k6, benchmarks
QA-004 Security Tester DAST, SAST, vulnerability scanning
QA-005 Manual QA Exploratory testing, UAT, regression
QA-006 Test Data Manager Fixtures, test environments, seeding


🎮 Commands

Command Description
/hivemind [task] Full multi-agent orchestration
/dev [task] Route to Development team
/sec [task] Route to Security team
/infra [task] Route to Infrastructure team
/qa [task] Route to QA team
/architect [task] Direct to DEV-001
/pentest [task] Direct to SEC-002
/sre [task] Direct to INF-005
/reviewer [task] Direct to DEV-004
/status System status
/recall [query] Query memory


🎬 Live Example

Input:

Build a REST API with JWT authentication and deploy to Kubernetes

Agent Execution:

[DEV-001] Designing architecture
[SEC-001] Threat modeling auth
[DEV-002] Building endpoints
[DEV-004] Reviewing code
[SEC-002] Testing security
[QA-002] Writing tests
[QA-003] Load testing
[INF-001] K8s architecture
[INF-005] Configuring deploy
[INF-006] Writing Helm charts

Quality Gates:

[GATE] G1-DESIGN .... PASSED
[GATE] G2-SECURITY .. PASSED
[GATE] G3-CODE ...... PASSED
[GATE] G4-TEST ...... PASSED
[GATE] G5-DEPLOY .... READY

Execution Report:

╔════════════════════════════════════════════════╗
║          HIVEMIND EXECUTION REPORT             ║
╠════════════════════════════════════════════════╣
║  Task: REST API + JWT + K8s Deploy             ║
║  Status: COMPLETE                              ║
╠════════════════════════════════════════════════╣
║  AGENTS: 11 engaged                            ║
║  ────────────────────────────────              ║
║  DEV-001 Architect .......... Complete         ║
║  SEC-001 Security Arch ...... Complete         ║
║  DEV-002 Backend Dev ........ Complete         ║
║  DEV-004 Code Reviewer ...... Complete         ║
║  SEC-002 Pentester .......... Complete         ║
║  QA-002  Test Automation .... Complete         ║
║  QA-003  Performance ........ Complete         ║
║  INF-001 Infra Architect .... Complete         ║
║  INF-005 SRE ................ Complete         ║
║  INF-006 Automation ......... Complete         ║
╠════════════════════════════════════════════════╣
║  DELIVERABLES:                                 ║
║  • REST API with CRUD endpoints                ║
║  • JWT auth + refresh tokens                   ║
║  • RBAC implementation                         ║
║  • Test suite (94 tests, 100% pass)            ║
║  • Security report (0 critical)                ║
║  • K8s manifests + Helm chart                  ║
║  • CI/CD pipeline                              ║
╠════════════════════════════════════════════════╣
║  Load tested at 8,000 req/sec                  ║
║  Auto-scaling configured                       ║
╚════════════════════════════════════════════════╝


🚦 Quality Gates

Every task passes through validation before completion:

┌────────┐   ┌────────┐   ┌────────┐   ┌────────┐   ┌────────┐
│   G1   │ → │   G2   │ → │   G3   │ → │   G4   │ → │   G5   │
│ DESIGN │   │SECURITY│   │  CODE  │   │  TEST  │   │ DEPLOY │
└────────┘   └────────┘   └────────┘   └────────┘   └────────┘
     │            │            │            │            │
     ▼            ▼            ▼            ▼            ▼
  Arch OK     No vulns     Review OK   Tests pass   Ready


🔀 Execution Patterns

Parallel Execution

Independent tasks run simultaneously

  DEV-001      SEC-001      INF-001      QA-001
     │            │            │            │
     ▼            ▼            ▼            ▼
┌─────────┐ ┌─────────┐ ┌─────────┐ ┌─────────┐
│Designing│ │Modeling │ │Planning │ │Strategy │
└────┬────┘ └────┬────┘ └────┬────┘ └────┬────┘
     └───────────┴───────────┴───────────┘
                      │
                      ▼
                 [ COMBINE ]

Sequential Execution

Dependent tasks chain together

┌─────────┐   ┌─────────┐   ┌─────────┐   ┌─────────┐
│ DEV-002 │ → │ DEV-004 │ → │  QA-002 │ → │ INF-005 │
│ Building│   │Reviewing│   │ Testing │   │Deploying│
└─────────┘   └─────────┘   └─────────┘   └─────────┘


📦 Installation

Prerequisites

Requirement Version
Python 3.11+
Node.js 18+
Codex CLI npm install -g @openai/codex
Claude CLI npm install -g @anthropic-ai/claude-code

Commands

# Install
./install.sh

# Run
hivemind

# Uninstall
./uninstall.sh


⚙️ Configuration

Environment Variables

export OPENAI_API_KEY="sk-..."
export ANTHROPIC_API_KEY="sk-ant-..."
export HIVEMIND_THEME="cyberpunk-matrix"

Config Files

config/
├── hivemind.yaml    # Main config
├── agents.json      # Agent definitions
├── routing.json     # Task routing
└── engines.yaml     # AI engines


⌨️ Keyboard Shortcuts

Key Action
Enter Send message
C Full chat mode
M Main view
Q Quit
Ctrl+O Status log
Ctrl+C Cancel task
Ctrl+Enter Send (chat)
Ctrl+L Clear history
Esc Back


📁 Project Structure

HIVEMIND/
├── agents/        # Agent definitions (24 specs)
├── comms/         # Inter-agent protocols
├── config/        # System configuration
├── core/          # Orchestrator logic
├── memory/        # Persistent memory
├── protocols/     # Quality gates
├── runtime/       # Execution controllers
├── teams/         # Team configurations
├── templates/     # Output templates
├── tests/         # Test suite
├── tui/           # Terminal UI
├── workflows/     # Process playbooks
├── hivemind       # Main executable
└── CLAUDE.md      # System prompt


🧠 Memory System

HIVEMIND remembers across sessions.

memory/
├── short-term/    # Session (clears on exit)
├── long-term/     # Permanent storage
│   ├── learnings
│   ├── preferences
│   ├── decisions
│   └── project
└── episodic/      # Events & milestones

Triggers

You say... HIVEMIND stores...
"Remember that..." → learnings
"We decided..." → decisions
"I prefer..." → preferences
"Our stack is..." → project


📄 License

MIT License — see LICENSE



🐝 HIVEMIND

Orchestrate Silently. Report Completely.


⬆ Back to Top

About

Engine-agnostic multi-agent orchestration system for AI coding assistants. 24 specialized agents across 4 teams (Development, Security, Infrastructure, QA) with CLI runtime, policy-based routing, persistent memory, and workflow automation. Works with Codex and Claude Code.

Resources

Readme

License

MIT license

Contributing

Contributing
Activity

Stars

1 star

Watchers

0 watching

Forks

1 fork
Report repository

Releases

No releases published

Sponsor this project

  •  
Learn more about GitHub Sponsors

Packages

 
 
 

Contributors

Languages