AgentRisk M2M 🛡️
Pre-Trade Security Layer for Autonomous DeFi Agents on Base.Stop letting your autonomous trading bots get rekt by stealth honeypots, malicious taxes, and rugpulls.
⚡ The Problem
Autonomous trading bots are fast, but they are completely blind.Every minute, hundreds of unvetted tokens launch on Base. Over 90% are engineered traps designed to block sells, drain executing wallets, or alter transfer taxes at the worst possible moment.Static blocklists (like standard free APIs) are useless against freshly deployed, obfuscated contracts. By the time human-curated lists update, your bot's capital is already gone.
🎯 The Solution
AgentRisk is an isolated, machine-readable security API built specifically for AI agents and automated scripts.Before your bot executes a swap() on Uniswap or Aerodrome, it pings AgentRisk. We run a deep runtime simulation, trace the deployer's dirty history, and return a strict, binary verdict:
{
"status": "success",
"token": "0xYourTargetToken...",
"recommendation": "REJECT",
"confidence": 0.99,
"reasons": [
"Honeypot detected: sell simulation failed",
"Deployer address linked to multiple previous rugpulls"
]
}⚙️ How It Works (M2M Architecture)
Agent Discovers Token via mempool or DEX router event.
Agent Calls AgentRisk MCP Tool (check_token_risk).
Instant x402 Micropayment ($0.15 USDC instantly settled on Base — no cumbersome API keys, subscriptions, or credit cards; pure machine-to-machine payment).
Binary Decision: The agent receives ALLOW, CAUTION, or REJECT. If it's not ALLOW, the bot aborts the trade instantly.
🚀 Quick Integration
Option 1: MCP Server (For Claude, Cursor & Custom Agents)
Add AgentRisk to your MCP config:
{
"mcpServers": {
"agentrisk": {
"command": "python",
"args": ["-m", "agentrisk_m2m.server"]
}
}
}
Option 2: Direct Python SDK / API Call
from agentrisk import AgentRiskClient
client = AgentRiskClient(endpoint="[https://agentrisk.dev](https://agentrisk.dev)", network="base")
# Pre-trade check before executing swap
verdict = client.check_token(
token_address="0xTargetTokenAddress...",
amount_usdc=50
)
if verdict.recommendation == "ALLOW":
execute_swap()
else:
print(f"Trade aborted. Risk detected: {verdict.reasons}")
💰 Economy
Model: Pay-per-scan via the x402 protocol.
Cost: 0.15 USDC per comprehensive deep simulation.
ROI: Losing $100 on a single honeypot costs more than 600 security scans. Protect your liquidity.
Built for autonomous systems that trust math, not hype.
