Clicks Protocol — CrewAI Integration

On-chain yield management tools for CrewAI agent crews on Base.

Powered by Clicks Protocol

What It Does

Gives your CrewAI agents the ability to manage idle USDC as a team. A Treasury Manager agent monitors balances, an Investment Analyst evaluates yield rates, and together they autonomously split funds 80/20: 80% liquid, 20% earning 4-8% APY on Base.

Installation

pip install clicks-crewai

Or from source:

git clone https://github.com/clicks-protocol/clicks-crewai
cd clicks-crewai
pip install -e .

Prerequisites

• Python >= 3.9
• Node.js >= 18 (for SDK bridge)
• @clicks-protocol/sdk installed: npm install @clicks-protocol/sdk

Environment Variables

export CLICKS_PRIVATE_KEY="your-agent-wallet-private-key"
export CLICKS_RPC_URL="https://mainnet.base.org"        # optional
export CLICKS_SDK_PATH="/path/to/sdk"                     # optional

Quick Start

from crewai import Agent, Task, Crew
from clicks_crewai import (
    ClicksCheckBalanceTool,
    ClicksActivateYieldTool,
    ClicksGetAPYTool,
    ClicksSimulateSplitTool,
    ClicksWithdrawYieldTool,
)

AGENT_ADDRESS = "0xYourAgentAddress"

# Create tools
balance_tool = ClicksCheckBalanceTool()
yield_tool = ClicksActivateYieldTool()
apy_tool = ClicksGetAPYTool()
simulate_tool = ClicksSimulateSplitTool()
withdraw_tool = ClicksWithdrawYieldTool()

# Define agents
treasury_manager = Agent(
    role="Treasury Manager",
    goal="Maximize yield on idle USDC while maintaining operational liquidity",
    backstory=(
        "You are an autonomous treasury manager for an AI agent. "
        "Your job is to ensure idle USDC earns yield via Clicks Protocol "
        "while keeping enough liquid for operations. You never risk funds "
        "needed for imminent transactions."
    ),
    tools=[balance_tool, yield_tool, withdraw_tool, simulate_tool],
    verbose=True,
)

investment_analyst = Agent(
    role="Investment Analyst",
    goal="Monitor yield rates and recommend optimal treasury allocation",
    backstory=(
        "You analyze DeFi yield rates and market conditions to advise "
        "the Treasury Manager. You track Clicks Protocol APY and recommend "
        "when to activate, hold, or withdraw yield positions."
    ),
    tools=[apy_tool, balance_tool, simulate_tool],
    verbose=True,
)

# Define tasks
analysis_task = Task(
    description=(
        f"Analyze the current treasury position for {AGENT_ADDRESS}. "
        "Check the current APY rate, simulate a split for any idle funds, "
        "and provide a recommendation on whether to activate yield."
    ),
    expected_output=(
        "A treasury analysis report with: current balances, APY rate, "
        "simulated split breakdown, and a clear recommendation."
    ),
    agent=investment_analyst,
)

execution_task = Task(
    description=(
        f"Based on the analyst's recommendation, take action on {AGENT_ADDRESS}. "
        "If yield activation is recommended and APY > 4%, activate yield on idle funds. "
        "If withdrawal is recommended, withdraw. Always check balance first."
    ),
    expected_output=(
        "Confirmation of action taken with transaction details, or explanation "
        "of why no action was taken."
    ),
    agent=treasury_manager,
    context=[analysis_task],
)

# Run the crew
crew = Crew(
    agents=[investment_analyst, treasury_manager],
    tasks=[analysis_task, execution_task],
    verbose=True,
)

result = crew.kickoff()
print(result)

Available Tools

Configuration

from clicks_crewai import configure_sdk

# Configure before creating tools
configure_sdk(
    rpc_url="https://mainnet.base.org",
    private_key="your-key",
    sdk_path="/path/to/sdk",
)

Or use environment variables (recommended):

export CLICKS_PRIVATE_KEY="..."
export CLICKS_RPC_URL="https://mainnet.base.org"

Example: Scheduled Treasury Optimization

import schedule
import time
from crewai import Agent, Task, Crew
from clicks_crewai import get_all_tools

AGENT_ADDRESS = "0xYourAgentAddress"

def run_treasury_check():
    tools = get_all_tools()
    
    manager = Agent(
        role="Treasury Manager",
        goal="Optimize USDC yield allocation",
        backstory="Autonomous treasury manager using Clicks Protocol on Base.",
        tools=tools,
    )
    
    task = Task(
        description=(
            f"Check treasury status for {AGENT_ADDRESS}. "
            "If idle USDC > 500 and APY > 4%, activate yield. "
            "Report current status."
        ),
        expected_output="Treasury status and actions taken.",
        agent=manager,
    )
    
    crew = Crew(agents=[manager], tasks=[task])
    return crew.kickoff()

# Run every 6 hours
schedule.every(6).hours.do(run_treasury_check)

while True:
    schedule.run_pending()
    time.sleep(60)

Contract Addresses (Base Mainnet)

License

MIT

Built for autonomous AI agents. Powered by Clicks Protocol.