moxxy

moxxySelf-hosted Multi-Agent AI Framework

Run local AI agents with isolated memory, tool execution, and multi-channel integrations

Get Started
View on GitHub
Moxxy
🤖

Multi-Agent Isolation

Each agent has its own SQLite memory, persona, encrypted vault, and extensible skills.

🧠

ReAct Loop Execution

Agents reason and act using LLM-driven tool invocation with a transparent execution model.

📱

Multiple Interfaces

Web dashboard, terminal UI, Telegram, Discord, Slack, and WhatsApp integrations.

🔧

Extensible Skills

Shell scripts, Python, MCP servers, or custom tools - extend agents with any capability.

🔒

WASM Sandboxing

Optional containerized execution with capability-based permissions for secure operations.

⏰

Scheduled Autonomy

Cron-based heartbeats enable proactive agent behavior without user interaction.

📊

Swarm Intelligence

Agents share facts through a global knowledge base for collaborative problem-solving.

📦

Self-Hosted

Complete data sovereignty. Your agents, your infrastructure, your rules.

Quick Start ​

Get up and running in minutes:

bash
# Install moxxy
curl -fsSL https://raw.githubusercontent.com/moxxy-ai/moxxy/main/install.sh | bash

# Run first-time setup
moxxy init

# Start the web dashboard
moxxy gateway start
moxxy web

Architecture Overview ​

┌──────────────────────────────────────────────────┐
│                  moxxy Gateway                   │
├──────────┬───────────┬───────────┬───────────────┤
│ Telegram │  Discord  │   Slack   │   WhatsApp    │
│   Bot    │   Bot     │   Bot     │   Bridge      │
├──────────┴───────────┴───────────┴───────────────┤
│              AutonomousBrain (ReAct)              │
│         LLM → <invoke> → Skill → Result          │
├──────────────────────────────────────────────────┤
│  Agent 1          │  Agent 2          │  Agent N  │
│  ┌─────────────┐  │  ┌─────────────┐  │          │
│  │ memory.db   │  │  │ memory.db   │  │   ...    │
│  │ persona.md  │  │  │ persona.md  │  │          │
│  │ vault       │  │  │ vault       │  │          │
│  │ skills/     │  │  │ skills/     │  │          │
│  └─────────────┘  │  └─────────────┘  │          │
├──────────────────────────────────────────────────┤
│              Shared swarm.db                     │
└──────────────────────────────────────────────────┘

What Can Moxxy Do? ​

  • Automate repetitive tasks - Let agents handle emails, reports, and data processing
  • Monitor systems - Set up agents to watch logs, APIs, and send alerts
  • Build custom workflows - Chain skills together for complex operations
  • Create conversational bots - Deploy agents to Telegram, Discord, or Slack
  • Develop AI applications - Use the REST API to integrate AI into your apps

Documentation Sections ​

SectionDescription
Getting StartedInstallation and first steps
CLI ReferenceAll CLI commands and options
Core ConceptsAgents, skills, memory, and the brain
ConfigurationLLM providers, vault, and container setup
ChannelsConnect to Telegram, Discord, Slack, WhatsApp
SkillsBuilt-in skills and creating custom ones
API ReferenceREST API documentation
AdvancedWASM sandboxing, dev mode, webhooks

Community ​

License ​

Moxxy is open source under the MIT License.

Open source · Self-hosted · Data sovereign

© 2026 Moxxy. Built with Rust.