Intro To Agents Quickstart 9 15

Intro to Agents Quickstart

A collection of autonomous agents built using the OODA (Observe, Orient, Decide, Act) loop pattern with LM Studio integration.

🚀 NEW: Multi-Agent Communication via NATS

Agents can now communicate with each other using NATS messaging - like Slack for AI agents! See **[README_NATS.md](README_NATS.md)** for details.

**Features**:

• 📢 Broadcast announcements to all agents
• 💬 Direct messages between specific agents
• 🤝 Request/response patterns with timeouts
• 🔄 Task handoff between agents
• 💓 Automatic agent discovery via heartbeats
• 📊 Channel-based routing (agents.all, agents.direct.{name}, etc.)

**Quick Start NATS**:

# Start NATS server
docker run -p 4222:4222 -p 8222:8222 nats:latest

# Run multi-agent demo (3 agents)
python demo_nats_agents.py

# Or run Trip Planner & Weather Bot example (2 agents coordinating)
python demo_trip_weather_interaction.py

**Example: Trip Planner coordinating with Weather Bot** 🌤️ ✈️ See `TRIP_PLANNER_README.md` for a complete example of two agents working together!

• Trip-Planner agent plans day trips
• Weather-Bot provides weather information
• Trip-Planner requests weather from Weather-Bot via NATS
• Shows real multi-agent coordination

Agents Available

1. NATS-Enabled OODA Agent 🌐

Autonomous agents with inter-agent communication capabilities via NATS.

**Location**: `nats_ooda_agent.py`, `nats_agent_mixin.py`, `nats_config.py` **Features**:

• Full OODA loop autonomy
• NATS messaging for multi-agent coordination
• Direct messaging and broadcasting
• Request/response patterns
• Task handoff between agents
• Agent discovery and heartbeats

2. First Mate Agent

An enhanced autonomous agent with memory, context management, and tool calling capabilities.

**Location**: `first_mate_agent/` **Features**:

• Context consolidation for long conversations
• Tool calling support
• Centralized logging
• Configuration management

3. Book Writer Agent 📚

An autonomous agent