Open-source reference implementations for AI-enabled payment security, blockchain fraud detection, and AML/CFT compliance reasoning.
Complete reverse engineering of Alipay SecurityGuard SDK — 9 CVEs (MITRE #2005801), AVMP VM bypass, 396/408 (97%) unprotected JSBridge APIs
Trace-backed Agent Authorization Reviews for tool-using AI agents. Staging-only evidence for payment, record-change, access, and export actions.
The Unified Payments Interface (UPI) has revolutionized the way people make payments in India. However, with the increasing adoption of UPI, there has been a rise in fraudulent transactions. To combat this, a UPI fraud detection system Project Includes Source Code, PPT, Synopsis, Report, Documents, Base Research Paper & Video tutorials
AI agent skill atlas for safer Arab/MENA payment and BNPL integrations.
AI Agent 付款安全层: 身份/意图/密钥 三道信任支柱 + 四道闸门。给 AI 定义带消费上限的数字身份(Spend-Capped Identity), 防提示注入, 密钥不落盘。Python + MCP。
Guardrails for Claude Code and OpenAI Codex reviewing payment systems: idempotency, ledgers, webhooks, reconciliation, and state machines.
The complete payment cryptography toolkit — 45+ tools for encryption, hashing, key management, PIN operations, and an interactive Thales payShield 10K HSM Simulator.
PayShield is a fraud detection and secure payment monitoring platform built with Cloudflare Workers, D1, KV, TypeScript, and a web-based dashboard.
Find payment logic flaws in Razorpay and Stripe checkouts - client-trusted amounts, rupee/paise bugs, unverified webhooks
Deterministic verification gate for agent-initiated payments. No model on the money path.
Enterprise-grade PCI-DSS v4.0 automated testing framework: API, UI & security test coverage with a built-in mock payment server. TypeScript + Jest + Playwright. Plug-and-play template for QA engineers.
Evidence firewall for autonomous AI payments, built with Jac and a deterministic Guardian policy engine.
Local-first pre-action / pre-settlement trust enforcement layer for agent-
Machine learning-based real-time UPI fraud detection system using behavioral analysis, anomaly detection, NLP, and multiple intelligent ML models.
PayFence Is A Payment-Control Layer For AI Agents That Need To Interact With Paid APIs And Services Autonomously. .
Deterministic authority and consumer execution for autonomous agents. Policy checks, exact approvals, x402 settlement, delivery verification and X Layer receipts.
Independent, citeable reference to the PCI security standards landscape - version lineage, status, sunset dates, and cross-family relationships. Not affiliated with or endorsed by PCI SSC.
A comprehensive machine learning project for detecting fraudulent payment transactions. Features 50,000 synthetic transactions, 4 ML models (Random Forest, XGBoost, LightGBM, CatBoost), 8 data visualizations, and SMOTE for handling class imbalance. Perfect for financial institutions and data scientists working on fraud detection systems.
Stop card testing, bot traffic, and fake signups on any site. Device identity that survives cleared cookies and rotating IPs, velocity across device/subnet/ASN/BIN, and a reason code behind every verdict. Drop-in script tag, self-hosted, MIT.
To associate your repository with the payment-security topic, visit your repo's landing page and select "manage topics."