from agentic_fleet import Agent

# Create simple agent
agent = Agent(
    name="assistant",
    model="gpt-4"
)

# Create agent with configuration
from agentic_fleet import AgentConfig

config = AgentConfig(
    name="researcher",
    model="gpt-4",
    temperature=0.7,
    max_tokens=2000,
    tools=["web_search", "file_reader"]
)

agent = Agent(config)

from agentic_fleet.agents import ChatAgent

agent = ChatAgent(
    name="chat_assistant",
    personality="helpful and friendly",
    context_window=10
)

# Chat with agent
response = await agent.chat(
    "Tell me about AgenticFleet"
)

from agentic_fleet.agents import TaskAgent

agent = TaskAgent(
    name="task_executor",
    tools=["file_reader", "code_executor"]
)

# Execute task
result = await agent.execute_task(
    "Analyze this Python file"
)

from agentic_fleet.agents import ResearchAgent

agent = ResearchAgent(
    name="researcher",
    search_depth=2,
    max_sources=5
)

# Conduct research
findings = await agent.research(
    "Latest developments in AI"
)

from agentic_fleet.memory import Memory

# Create memory
memory = Memory(
    storage_type="redis",
    ttl=3600
)

# Create agent with memory
agent = Agent(
    name="assistant",
    memory=memory
)

# Store information
await agent.remember(
    key="user_preference",
    value="prefers detailed explanations"
)

# Recall information
preference = await agent.recall(
    "user_preference"
)

from agentic_fleet.tools import WebSearch, FileReader

# Add tools to agent
agent.add_tools([
    WebSearch(),
    FileReader()
])

# Use tool
result = await agent.use_tool(
    "web_search",
    query="AgenticFleet documentation"
)

# Send message
response = await agent.send(
    "Hello, can you help me?"
)

# Structured communication
response = await agent.communicate(
    intent="greeting",
    content="Hello",
    context={"user": "John"}
)

from agentic_fleet.agents import AgentGroup

# Create agent group
group = AgentGroup([
    Agent(name="researcher"),
    Agent(name="writer")
])

# Group communication
result = await group.collaborate(
    task="Write a research paper"
)

# Save agent state
state = await agent.save_state()

# Load agent state
new_agent = Agent.load_state(state)

# Reset agent
await agent.reset()

from agentic_fleet.monitoring import AgentMonitor

monitor = AgentMonitor(agent)

# Track performance
await monitor.track_performance()

# Get metrics
metrics = await monitor.get_metrics()

from agentic_fleet.exceptions import AgentError

try:
    response = await agent.process(
        "complex task"
    )
except AgentError as e:
    print(f"Agent error: {e}")

from agentic_fleet.behaviors import Behavior

class CustomBehavior(Behavior):
    async def apply(self, agent):
        # Implement behavior
        pass

agent.add_behavior(CustomBehavior())

from agentic_fleet.learning import Learning

learning = Learning(
    strategy="reinforcement",
    model="gpt-4"
)

agent.enable_learning(learning)

from agentic_fleet.optimization import optimize_agent

optimized_agent = await optimize_agent(
    agent,
    metric="response_time"
)

from agentic_fleet.deployment import deploy_agent

# Deploy agent
deployment = await deploy_agent(
    agent,
    platform="kubernetes",
    replicas=3
)

# Scale deployment
await deployment.scale(replicas=5)