GuardClaw

MERCHANTGUARD™

The Compliance Layer for Payments & AI Agents

Install once. Compliance everywhere.

Quick Start
claude mcp add guard --transport sse https://merchantguard-mcp-810654658669.us-central1.run.app/sse
One command — 25 tools instantly available in Claude Code
1.Enter your business details
2.Get your risk score (0-100)
3.Follow your personalized action plan

Don't have an AI agent? Get certified

Get alerts when Visa changes chargeback rules

I agree to receive email updates and accept the Privacy Policy

760+
npm Downloads
24+
Processors Matched
14
Verticals Covered
0.9%
VAMP Threshold
Listed in the Coinbase x402 EcosystemCompliance layer for agent payments
Stripe x402 Compatible
Chrome AI Early Preview
Base Mainnet
10 Probe Types
24+ PSPs

Everything You Need to Stay Off the MATCH List.

Visa fines merchants $25K+ for high chargebacks. Guard™ scores your risk, shows you exactly what to fix, and finds processors that will approve your business.

0-100

Risk Score

Know your chargeback risk before Visa does. Upload your processor CSV or answer 14 questions — get your score in 60 seconds.

3x

Security Scanner

Scan your payment setup for vulnerabilities. 102 checks across PCI compliance, fraud prevention, and processor requirements.

13

Guard™ Covers 14 Industries

One AI coach that knows YOUR vertical. CBD, crypto, supplements, adult, gaming, travel — deep compliance rules for each.

50+

Chargeback Alerts

Get notified before you cross Visa's threshold. Real-time alerts on rule changes, processor policies, and VAMP enforcement.

10

Agent Certification

10-probe adversarial audit for AI agents. Security, ethics, reliability, PII handling. Earn Diamond/Gold/Verified tier.

24/7

Watchdog Monitoring

Free 24/7 monitoring for your payment agents. Health checks, suspension alerts, quality scoring.

Generic AI Gives Generic Advice. Guard™ Knows YOUR Numbers.

Guard™ remembers your industry, volume, chargeback rate, and processor. Every answer is about YOUR business, not a textbook.

ChatGPT / Generic AI

Here's how to reduce chargebacks: implement RDR, sign up for Verifi or Ethoca alerts, enable 3D Secure...

Guard™ (Knows YOUR Business)

You're processing $85K/month in gaming with 2.3% chargebacks. That's above Visa's 0.9% threshold. RDR will block 284 disputes/month, saving you $21,584 and keeping you off the MATCH list.

Volume$85K
CB Rate2.3%
IndustryGaming
RegionBrazil
PSPStripe
RiskVAMP Exposure

What Guard™ Knows That ChatGPT Doesn't

VAMP Formula
ChatGPT

Hallucinated or outdated thresholds

Guard™

Exact formula: (fraud×2 + chargebacks) ÷ settled CNP. Updated April 2026 enforcement rules.

Processor Thresholds
ChatGPT

Doesn't know Stripe bans at 0.65%

Guard™

Shadow thresholds for Stripe (0.65%), Adyen (0.4%), Square (0.55%), PayPal (1.0%) — processors kill you before Visa does.

PSP Matching
ChatGPT

Names 3-4 popular processors

Guard™

24+ processors with real approval profiles, verticals, reserve policies, and timelines. Matched to YOUR industry and volume.

Industry Rules
ChatGPT

Generic compliance advice

Guard™

CBD requires COA + <0.3% THC. Nutra can't make health claims. Subscriptions need FTC Click-to-Cancel. We know each vertical's rules.

Your Business
ChatGPT

Starts from zero every conversation

Guard™

Remembers your chargeback rate, volume, processor, industry, and region. Every answer builds on your data.

2,800+ hours of vertical compliance research. Can't be replicated with a prompt.

Works with your processor

Stripe logo
PayPal logo
Shopify logo
Square logo
PayKings logo
PaymentCloud logo
NMI logo
Worldpay logo
Paysafe logo
Stripe logo
PayPal logo
Shopify logo
Square logo
PayKings logo
PaymentCloud logo
NMI logo
Worldpay logo
Paysafe logo
Stripe logo
PayPal logo
Shopify logo
Square logo
PayKings logo
PaymentCloud logo
NMI logo
Worldpay logo
Paysafe logo

Also: Adyen, Authorize.Net, Braintree, Checkout.com, USAePay + 47 more

FREE

Free Agent Monitoring

Your agent goes down at 3am. Guard™ catches it. Suspension incoming? Guard™ alerts you. Quality dropping? Guard™ flags it. Free tier monitors 2 bots.

Health Pings
Suspension Alerts
Quality Scoring
Loop Detection
Karma Tracking
Telegram + Email
Start Free Monitoring

Free to Start. Pay As You Grow.

$0

Free Forever

Risk score, security scan, Guard™ AI coach (14 verticals), chargeback alerts, and on-chain compliance identity. No credit card.

$1=100

Pay Per Use

$1 = 100 credits. First 1,000 free. Each coach question = 1 credit. Each alert = 0.1 credit.

USDC

Agent Payments (x402)

AI agents pay per-call with USDC on Base. No API key needed. No subscription.

x402 USDC Per-Call Pricing
Alert feed pull$0.001
Coach query$0.01
GuardScan (pattern)$0.05
GuardScan (deep + AI)$0.15
Mystery Shopper probe$0.05
GuardScore$0.10
Full certification$0.50
BASE MAINNET

On-Chain Compliance Identity

Free soulbound NFT identity for every agent. GuardScorePassport for merchants. Verifiable on-chain.

GuardScorePassport0x94Ab...8741
ERC-8004 Identity/.well-known/erc8004.json

FAQ

What happens if my chargebacks are too high?
Visa's VAMP program fines merchants $25K+ when chargebacks exceed 0.9%. Guard™ scores your risk, shows exactly what to fix, and matches you with processors built for your industry — before Visa flags you.
How much does it cost?
Core features are 100% free forever: risk score, security scan, Guard™ AI coach (covers 14 industries), chargeback alerts, on-chain identity. No credit card. Premium audits start at $4.99.
What makes Guard™ different from ChatGPT?
Guard™ has 2,800+ hours of vertical compliance research that no AI model has in training data. It knows the real VAMP formula, every processor's secret termination thresholds, 24+ PSP approval profiles, and industry-specific regulations. Plus it remembers YOUR business — chargeback rate, volume, processor, industry. ChatGPT gives textbook answers. Guard™ gives you exact numbers for YOUR situation.
I'm an AI agent developer. How does Guard™ help?
Guard™ certifies your agent's compliance with a 10-probe adversarial audit. Agents pay per-call with USDC on Base via x402 — no subscriptions, no API keys. One install and your agent is compliance-ready.
Ask Guard™ AI
Add to Claude Code

MerchantGuard MCP

GuardScore, GuardScan, Mystery Shopper, 14 compliance coaches, agent certification — 25 tools in one command.

$claude mcp add merchantguard --transport sse https://merchantguard-mcp-810654658669.us-central1.run.app/sse
or install via npm: npm i @merchantguard/guard

Install Guard™. Get Protected.

One install. MCP, npm, or API. Guard™ learns your business and handles the rest.

Get Started Free

No credit card. No signup required for API.