Autonomous Compliance & Transaction Screening¶

The Problem With Manual Compliance¶

Traditional DeFi compliance is a contradiction in terms. Protocols are designed to operate 24/7 with no intermediaries — but compliance processes require humans to review wallets, file reports, and apply judgment to edge cases. The result is either under-compliance (protocols ignore the problem) or over-compliance (they implement clunky KYC gates that destroy the user experience).

There is a third path: autonomous compliance agents that screen wallets and transactions in real time, apply risk-based logic, and pass only the hard cases to human review — at zero marginal cost per decision.

What Autonomous Compliance Covers¶

ChainAware's agent-based compliance stack covers the full screening workflow without human intervention in the critical path:

Wallet Onboarding Screening¶

Every wallet connecting to your protocol is screened before access is granted:
- Fraud probability (98% accuracy on Ethereum)
- AML score — sanctions lists, mixer interactions, darknet exposure
- Behavioural risk profile — experience level, risk willingness, counterparty history
- Verdict: PASS / Enhanced Due Diligence / REJECT

Approved wallets are onboarded instantly. EDD cases are flagged for human review. Rejections are blocked automatically with a timestamped audit record.

Real-Time Transaction Monitoring¶

Every transaction initiated through your protocol is screened before it executes:
- Sender and receiver screened simultaneously
- Contract address screened if applicable
- Composite risk score (0–100) returned pre-execution
- Action: ALLOW / FLAG / HOLD / BLOCK

Transactions are processed in under 100ms. Your pipeline never waits for compliance.

Continuous Portfolio Monitoring¶

Wallets already in your system are re-screened automatically as new on-chain data emerges:
- Risk score changes above configurable thresholds trigger alerts
- Wallets that passed onboarding but subsequently exhibit fraud patterns are flagged
- Full audit trail of each score change with timestamp and contributing factors

MiCA Coverage¶

For EU-regulated protocols, ChainAware covers approximately 70–75% of DeFi-applicable MiCA requirements autonomously:

MiCA Requirement	Autonomous Coverage
Sanctions screening	✅ Real-time, every transaction
AML behavioural monitoring	✅ Continuous, model-based
Fraud and bot detection	✅ 98% accuracy
Transaction risk scoring	✅ Pre-execution, every tx
Timestamped audit records	✅ Automatic, exportable
Travel Rule	✅ Not applicable to pure DeFi
PEP screening	⚠️ Add third-party vendor

The Travel Rule does not apply to pure DeFi — transactions between non-custodial wallets and smart contracts are not VASP-to-VASP transfers.

Full MiCA Compliance Guide →

The Compliance Agent Stack¶

`chainaware-compliance-screener`¶

Orchestrates four sub-agents into a structured Compliance Report. Accepts single wallets or batches. Returns verdict, risk rating, per-check results, and an explicit scope disclaimer for regulatory documentation.

Compliance check before we onboard this wallet: 0xApplicant... on Ethereum

MiCA screening for this counterparty: 0xPartner... on BNB Chain

Batch onboarding compliance for our IDO waitlist — flag anything above EDD: [list]

Example output:

COMPLIANCE REPORT — PASS
Address: 0xUser... | Network: Ethereum

Risk Rating: LOW
Fraud probability: 0.03  ✅
AML score: 96/100        ✅
Forensic flags: None     ✅
Behaviour: Veteran, 4 years on-chain

Verdict: Proceed with standard onboarding.
Audit record: 2026-03-16T09:14:22Z

`chainaware-transaction-monitor`¶

Real-time pipeline screening for every agent-initiated or user-initiated transaction. Returns a machine-actionable verdict before broadcast.

Risk score for this pending swap — sender: 0x... receiver: 0x... on Base

Monitor this bridge transaction before relay: from=0x... to=0x... value=50 ETH

Example output:

Composite Risk Score: 8/100 — ✅ ALLOW
Sender: Low risk (0.03) | Receiver: Low risk (0.02)
No intervention required.

`chainaware-aml-scorer`¶

Lightweight AML-only screening for high-volume pipelines where a single score per wallet is sufficient.

AML score for 0x... on Ethereum before we process their withdrawal

Two Integration Paths¶

Path 1: Prediction MCP — For AI Agent Pipelines¶

The natural fit for protocols already using AI agents. Add ChainAware's MCP server and your compliance agent can screen wallets, monitor transactions, and generate audit records in natural language — no fixed rules to maintain.

claude mcp add --transport sse chainaware-behavioral-prediction \
  https://prediction.mcp.chainaware.ai/sse --header "X-API-Key: YOUR_KEY"

Once connected, your agent can autonomously decide when to invoke compliance checks based on context — screening high-value transactions more thoroughly than routine ones, applying stricter thresholds for new wallets than established users.

Path 2: Google Tag Manager — For dApp Front-Ends¶

No backend changes required. Add the ChainAware GTM pixel to your existing container and wallet-connect events automatically trigger compliance screening. PASS / EDD / REJECT verdicts are returned to the dataLayer, and your front-end renders the appropriate UI state.

Deployment takes under an hour. Audit records are delivered to a compliance inbox via webhook.

Path 3: REST API — For Custom Integrations¶

Direct API access for any stack:

# Onboarding screen
GET /v1/fraud/{wallet}?chain=ethereum

# Transaction screen
POST /v1/monitor/transaction
{ "from": "0x...", "to": "0x...", "value": "...", "chain": "ethereum" }

Full documentation at swagger.chainaware.ai.

Compliance Workflows by Protocol Type¶

DeFi Lending Protocol¶

Wallet connects → compliance-screener runs at onboarding → PASS / EDD / REJECT
Loan request submitted → transaction-monitor screens in real time → ALLOW / HOLD / BLOCK
Ongoing: aml-scorer re-screens active borrowers monthly

DEX Front-End¶

Wallet connects → fraud-detector + aml-scorer run in parallel (< 100ms)
Swap initiated → transaction-monitor screens sender + receiver + contract
High-risk swaps routed to manual review queue; others proceed instantly

Token Launchpad¶

Registration opens → compliance-screener batch-screens all applicants
EDD-flagged applicants held for manual KYC supplement
Allocations distributed only to PASS-verified wallets
Post-launch: transaction-monitor watches for suspicious redistribution patterns

Why Predictive AI Outperforms Rule-Based Compliance¶

Dimension	Rule-Based Systems	ChainAware Predictive AI
False positive rate	30–70%	Under 2%
New wallet detection	Very limited	Strong behavioural matching
Adapts to new fraud patterns	Manual rule updates	Daily model retraining
DeFi native	No	Yes
Cost per decision	High (human review)	Near zero (automated)

A 30–70% false positive rate doesn't just inconvenience users — it creates compliance liability. Blocking legitimate users without documented cause is as much a regulatory risk as failing to block bad actors.