惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

V
V2EX
P
Proofpoint News Feed
D
DataBreaches.Net
C
Check Point Blog
L
LangChain Blog
量子位
美团技术团队
Vercel News
Vercel News
人人都是产品经理
人人都是产品经理
N
Netflix TechBlog - Medium
V
Visual Studio Blog
Microsoft Security Blog
Microsoft Security Blog
博客园 - 【当耐特】
MongoDB | Blog
MongoDB | Blog
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
Last Week in AI
Last Week in AI
The GitHub Blog
The GitHub Blog
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
U
Unit 42
腾讯CDC
M
MIT News - Artificial intelligence
Microsoft Azure Blog
Microsoft Azure Blog
Blog — PlanetScale
Blog — PlanetScale

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
MoltsPay 2.0: When Agentic Commerce Stops Caring Whether ...
Yaqing2023 · 2026-06-23 · via DEV Community

Yaqing2023

TL;DR: MoltsPay 2.0.0 (published 2026-06-20) adds its first fiat payment rail — Alipay (支付宝 AI 收), settling in CNY — alongside the existing crypto rails (Base, Polygon, Solana, BNB, Tempo, XRPL). The same skill can now quote a price in USDC and in CNY, and the agent routes by preference and capability. Crucially, nothing about the crypto rails changed: this is a backward-compatible, additive release. The reason it was a config change and not a rewrite is the same reason we wrote about before — MoltsPay is abstraction-first, not protocol-first.


Where We Left Off

In Why MoltsPay is Emerging as the Best Stack for Multi-Chain Agentic Commerce the argument was simple:

Agentic commerce assumes the opposite of what today's payment systems were built for. Today's rails assume a human in the loop, single-chain execution, and explicit wallet interaction. Agents need programmatic, cross-chain payments with no human and no wallet ceremony.

The answer there was the Universal Payment Protocol (UPP) — an abstraction layer that wraps heterogeneous protocols (x402, MPP, Solana Pay) behind one unified API, so the agent expresses intent and MoltsPay picks the optimal protocol per chain. The whole bet was that abstraction-first beats protocol-first in a fragmented world.

2.0 is that bet paying off in the place it matters most: a payment type that isn't a blockchain at all.

The Next Wall: Crypto-Only Is Still a Silo

Multi-chain solved fragmentation within crypto. But "pay any chain" is still "pay in crypto." For real-world agentic commerce that's a hard ceiling:

  1. Most real merchants settle in fiat. A merchant in mainland China prices a service in CNY, reconciles in CNY, and is regulated in CNY. Telling them "accept USDC on Base" is a non-starter, not a UX nit.
  2. Most real buyers' agents hold fiat-backed balances, not a gas-funded wallet on six chains.
  3. An agent shouldn't know or care which world a price lives in. "This service costs ¥6.99" and "this service costs $0.99 USDC" should be the same shape of decision.

A protocol-first stack hits this wall hard: x402 is an HTTP-native crypto settlement protocol. Bolting fiat onto it means a parallel, bespoke integration with its own SDK, its own signing, its own error model — exactly the fragmentation UPP was built to hide.

MoltsPay 2.0's Move: Fiat Is Just Another Rail

In MoltsPay, a "rail" is the unit of abstraction. Adding Alipay didn't mean a new protocol surface for developers — it meant implementing one more rail behind the same UPP boundary.

The Alipay rail (支付宝 AI 收) at a glance:

  • Rail / chain id"alipay", type: "fiat-rail"; sits at the same level as EVM/SVM chains
  • Currency — CNY (amount is yuan as a string, not cents)
  • Settlement protocol — Alipay AI 收, RSA2 (SHA256WithRSA) merchant-key signing
  • Server — 100% native TypeScript, holds the RSA2 private key
  • Client — Node CLI shells out to alipay-bot (auto-provisioned on install); browser not yet supported
  • Eligibility — Mainland-China merchants only (ICP + business license + Alipay Open Platform)
  • Backward compatibility — Zero changes to crypto rails

The developer-facing change is one config block, not a new mental model:

{
  "provider": {
    "name": "My AI Service",
    "wallet": "0xYourEVMAddress",        // crypto rails, unchanged
    "alipay": {                           // NEW: fiat rail
      "seller_id": "2088...",
      "app_id": "2021...",
      "private_key_path": "./cert/ALIPAY_PRIVATE_KEY.txt",
      "alipay_public_key_path": "./cert/ALIPAY_PUBLIC_KEY.txt",
      "sign_type": "RSA2"
    }
  },
  "services": [{
    "id": "text-to-video",
    "price": 0.99, "currency": "USDC",    // crypto price
    "alipay": { "price_cny": "6.99" }     // NEW: fiat price on the same service
  }]
}

Now the same skill can be paid in USDC on six chains or in CNY via Alipay. The agent chooses.

How It Stays Backward-Compatible: Dual-Emit 402

The wire-level trick is the same 402 Payment Required flow, but the server now dual-emits payment requirements:

  • X-Payment-Required — the x402 / UPP-aware header that newer agents read
  • Payment-Needed — the legacy header existing MoltsPay clients already understand
Agent ──request──▶ Server
Agent ◀─402 + {X-Payment-Required, Payment-Needed}── Server
        (offers: USDC on Base/Solana/BNB/…  AND  CNY via Alipay)
Agent ──picks a rail, pays──▶ Server ──settles──▶ result

Because both headers ship together, an old crypto-only client sees exactly what it saw in 1.x, while a 2.0 agent additionally sees the fiat option. No client was forced to upgrade. That is what "additive release" means in practice, and it's only cheap because the abstraction boundary already existed.

Why This Matters for Agentic Commerce

  1. Intent over instrument. The agent says "buy this," not "sign an EIP-712 transfer on BNB" or "call the Alipay gateway with an RSA2 signature." Crypto and fiat collapse into the same programmable decision.
  2. Real merchants can onboard. A mainland-China merchant accepts agent payments in their own currency and compliance regime — and, if they want, also accepts USDC from overseas agents, from one skill file.
  3. Routing becomes a feature, not a fork. Buyer prefers fiat? Settle Alipay. Seller only wants crypto? Settle USDC. The decision lives in MoltsPay, not in every agent's code.
  4. The hardest integration validated the architecture. Fiat doesn't have a chain, gas, or a wallet — it's the least "crypto-shaped" thing you could add. Adding it as a rail rather than a rewrite is the strongest evidence yet that abstraction-first was the right call.

Abstraction-First, Restated

The original article compared MoltsPay to protocol-first stacks and argued abstraction wins as the world fragments. 2.0 sharpens the claim:

A protocol-first stack is only as broad as its protocol. An abstraction-first stack is as broad as the rails you can implement behind it — and rails don't have to be blockchains.

Crypto fragmentation was the first proof. Crossing the crypto/fiat boundary with a config change — and breaking nothing — is the second.

Try It

npm install moltspay        # auto-provisions the alipay-bot CLI (best-effort;
                            # skip with MOLTSPAY_SKIP_CLI_INSTALL=1)
npm view moltspay version   # 2.0.0

  • Crypto rails: Base · Polygon · Solana · BNB · Tempo · XRPL — gasless, USDC
  • Fiat rail: Alipay (支付宝 AI 收) — CNY, mainland-China merchants
  • Overseas merchants / individual developers: use the USDC rails (no onboarding required)
  • Full config reference: docs/ALIPAY-RAIL.md

About MoltsPay

MoltsPay is open-source payment infrastructure for AI agents: HTTP-native, intent-based payments where agents pay each other — and now real merchants — for services without a human in the loop. As of 2.0.0, that spans six crypto chains and one fiat rail behind a single Universal Payment Protocol.

Compatibility note: 2.0.0 introduces no breaking changes to the existing crypto rails. Existing 1.x users can upgrade directly.

Tags: #ai #agenticcommerce #payments #alipay #fiat #crypto #x402 #web3 #opensource