​
Quickstart with OVADARE

This guide will help you set up OVADARE in minutes to detect and resolve conflicts in your multi-agent systems.

​
Prerequisites

Ensure you have the following:
  • Python 3.7 or higher installed.
  • A multi-agent platform like AutoGen integrated with your project.
  • OVADARE installed. If not, follow the Installation Guide.

​
Step 1: Import OVADARE into Your Project

Start by importing the required modules from OVADARE: 
from ovadare.agents.agent_registry import AgentRegistry
from ovadare.conflicts.conflict_detector import ConflictDetector
from ovadare.resolution.resolution_engine import ResolutionEngine

​
Step 2: Register Your Agents

Register the agents you want OVADARE to monitor. For example: 
agent_registry = AgentRegistry()

agent_registry.register_agent(
    agent_id="lead_researcher",
    capabilities={"analyze_data": True, "prepare_reports": True}
)

agent_registry.register_agent(
    agent_id="sales_strategist",
    capabilities={"create_sales_strategy": True, "evaluate_trends": True}
)

​
Step 3: Detect Conflicts

Set up the Conflict Detector to analyze actions and identify potential conflicts: 
conflict_detector = ConflictDetector(agent_registry)

conflicts = conflict_detector.detect(
    agent_id="lead_researcher",
    action={"task": "create_sales_strategy", "resource": "market_data"}
)

if conflicts:
    print("Conflicts detected:", conflicts)
else:
    print("No conflicts detected.")

​
Step 4: Resolve Conflicts

Use the Resolution Engine to resolve detected conflicts dynamically: 
resolution_engine = ResolutionEngine(agent_registry)

resolutions = resolution_engine.generate_resolutions(conflicts)

for resolution in resolutions:
    print("Resolution applied:", resolution)

​
Step 5: Iterate and Improve

Integrate feedback from conflict resolution to refine your policies and improve collaboration between agents. Explore how to define custom policies in the Policy Management Guide.

​
Explore More

Ready to dive deeper? Check out: