crewai-paylobster

CrewAI integration for PayLobster - AI agent payment escrow on Base.

Installation

pip install crewai-paylobster

Quick Start

from crewai import Agent, Task, Crew
from crewai_paylobster import (
    create_escrow_tool,
    release_escrow_tool,
    check_reputation_tool,
    register_agent_tool,
    get_balance_tool,
)

# Create a Payment Agent with PayLobster tools
payment_agent = Agent(
    role="Payment Manager",
    goal="Handle secure escrow payments for AI agent work",
    backstory="""You are an expert at managing cryptocurrency payments.
    You verify worker reputation, create secure escrows, and release
    payments when work is completed.""",
    tools=[
        create_escrow_tool,
        release_escrow_tool,
        check_reputation_tool,
        get_balance_tool,
    ],
    verbose=True,
)

# Create a task
payment_task = Task(
    description="""Check the reputation of worker 0x742d35Cc6634C0532925a3b844Bc9e7595f0bEb.
    If they have a good track record, create an escrow for $150 USDC to pay them
    for data analysis work, with a 48-hour deadline.""",
    agent=payment_agent,
    expected_output="Confirmation that reputation was checked and escrow was created",
)

# Create crew and run
crew = Crew(agents=[payment_agent], tasks=[payment_task])
result = crew.kickoff()
print(result)

Available Tools

create_escrow_tool

Create a payment escrow that locks USDC funds until work is completed.

release_escrow_tool

Release payment to the worker after work is completed.

check_reputation_tool

Check an agent's reputation score and task history.

register_agent_tool

Register your agent on the PayLobster identity system.

get_balance_tool

Check USDC balance of any address on Base.

Example: Payment Crew

See examples/payment_crew.py for a complete example with:

• A Payment Manager agent that handles all escrow operations
• A Negotiator agent that discusses terms with workers
• Coordination between agents for complete payment workflows

Contract Addresses (Base Mainnet)

• Identity: 0xA174ee274F870631B3c330a85EBCad74120BE662
• Reputation: 0x02bb4132a86134684976E2a52E43D59D89E64b29
• Escrow: 0x703B528C1b07cd27992af9Ae11DD67bE685E489e
• Credit: 0x4c22B52eacAB9eD2Ce018d032739a93eC68eD27a
• USDC: 0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913

Security

⚠️ Never hardcode private keys in production! Use environment variables:

import os
private_key = os.environ.get("AGENT_PRIVATE_KEY")

License

MIT