Ripple introduces XRPL toolkit to support AI agent transactions

Ripple has launched a new developer toolkit for building AI-powered payment applications on the XRP Ledger, adding support for x402 payments using XRP and RLUSD as interest in agentic transactions continues to grow.

According to Ripple, the newly released XRPL AI Starter Kit is designed to help developers build applications that allow AI agents to send, receive, and manage payments on the XRP Ledger with limited human involvement.

The company said the launch is being rolled out in phases, with the first stage focused on developer tools, documentation, and integrations.

In an announcement published Wednesday, Ripple said AI agents are already carrying out tasks such as paying for computing resources, settling invoices, and completing transactions. Existing payment systems were largely built around human approvals and reconciliation processes, Ripple said, while autonomous software requires infrastructure that can settle transactions quickly and operate without manual intervention.

The initial release includes access to the XRPL Docs MCP Server, which allows compatible clients such as Claude Code, Claude Desktop, Cursor, and custom agent frameworks to retrieve XRP Ledger documentation when needed. 

Developers can also use new wallet and payment skills for Claude that support functions including wallet creation, balance checks, payments, and transaction tracking.

XRP and RLUSD added to x402 payment standard

Alongside the toolkit launch, Ripple announced support for the x402 protocol on the XRP Ledger through a contribution from partner t54. The integration enables AI agents to use XRP or Ripple USD (RLUSD) to pay for API calls, AI model inference, and other internet-based services through the open payment standard.

Explaining why it sees the XRP Ledger as suitable infrastructure for AI-driven payments, Ripple pointed to deterministic transaction finality, settlement times of roughly 3 to 5 seconds, and predictable transaction costs that do not rely on gas fee auctions.

The company also highlighted the network’s native decentralized exchange and multi-currency payment capabilities. 

According to Ripple, an agent can complete a transaction that sends RLUSD while delivering XRP to a recipient, with the conversion handled through the protocol’s built-in exchange layer rather than external bridges or swap contracts.

Ripple further noted that the XRP Ledger has operated continuously since 2012 without transaction rollbacks. For organizations deploying AI agents that manage funds, the company said that operating history offers an additional layer of reliability.

Institutional controls such as escrow, multi-signing, deposit authorization, and trust lines were also cited as tools that can help organizations define spending permissions and transaction rules for agents without relying on custom smart contracts.

Meanwhile, RLUSD was positioned as a payment rail for agentic use cases that require price stability, including invoice settlement, payroll, and agent-to-agent commerce. Ripple said the stablecoin, which is native to XRPL, can be exchanged for other assets through the network’s decentralized exchange while allowing transactions to remain denominated in U.S. dollars.

The launch comes as technology and crypto companies continue to explore infrastructure for AI-driven financial activity. Robinhood recently introduced an initiative that allows users to experiment with AI agents trading equities, while MetaMask has released a non-custodial wallet designed for AI agents.

Not everyone agrees that AI systems can operate independently of people. Earlier this week, researchers from IC3, a consortium involving several leading universities, argued that while blockchain wallets can automate transactions and access on-chain services, AI agents remain dependent on human operators and the underlying infrastructure supporting them.

Ripple described the AI starter kit as the first phase of a longer development effort and said future releases will be shaped by developer feedback and emerging payment use cases.