-
Notifications
You must be signed in to change notification settings - Fork 18
Open
Description
Suggestion
Before a buyer agent creates a job with a seller, it could check the seller's counterparty risk (wallet history, cross-chain trust, sanctions screening).
Revettr provides this as a simple API call — score any wallet 0-100 before committing funds. Works with x402 on Base (same chain as ACP).
Example flow
from revettr import Revettr
from virtuals_acp.client import VirtualsACP
from virtuals_acp.models import ACPGraduationStatus, ACPOnlineStatus
# Standard ACP setup
acp_client = VirtualsACP(acp_contract_clients=...)
# Browse for sellers
agents = acp_client.browse_agents(
keyword="data analysis",
graduation_status=ACPGraduationStatus.ALL,
online_status=ACPOnlineStatus.ONLINE,
)
chosen_agent = agents[0]
# Score the seller BEFORE creating a job
revettr = Revettr()
risk = revettr.score(wallet_address=chosen_agent.wallet_address)
if risk.score < 60:
print(f"Seller scored {risk.score}/100 ({risk.tier}) — too risky, skipping")
else:
# Safe to create job
job_id = chosen_agent.job_offerings[0].initiate_job(
service_requirement={"task": "Analyze Q1 sales data"},
evaluator_address=evaluator_address,
)Available as
- Python SDK:
pip install revettr - MCP server:
pip install revettr[mcp]
Full example: https://github.com/AlexanderLawson17/revettr-python/blob/feat/virtuals-acp-integration/examples/virtuals_acp_safe_buyer.py
Not trying to sell anything — just flagging a useful safety layer for agent-to-agent commerce. Happy to help with integration if there's interest.
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
No labels