Key points
- Version 1.1 adds support for the Machine Payments Protocol and Open Wallet Standard.
- One-time payments can use XRP and XRPL-issued assets, while session payments currently rely on XRP payment channels.
- The software remains in beta, and broader asset support for payment channels depends on a proposed XRPL amendment.
RippleX has released version 1.1 of its XRP Ledger AI Starter Kit, adding support for two open specifications intended to help software agents make payments and interact with wallets under explicit controls. The update connects the kit to the Machine Payments Protocol, or MPP, and the Open Wallet Standard, or OWS. RippleX described the release in a developer post updated September 16, while CoinDesk reported the integration on September 17.
Payments through an HTTP workflow
MPP was co-authored by Stripe and Tempo as an open protocol for programmatic payments. Its basic flow uses the web's HTTP 402 Payment Required response: an agent requests a resource, receives a payment request, authorizes the charge and then gains access to the resource. Stripe says the design can support one-time charges, recurring payments and other machine-to-machine transactions without forcing every service to build a separate checkout process.
Related reporting: Ripple targets corporate treasury flows in next phase of RLUSD expansion
In the RippleX implementation, the XRP Ledger acts as one settlement option inside that flow. The starter kit can handle single-charge payments using XRP, issued currencies and multipurpose tokens. RippleX also lists RLUSD as an example of an XRPL-issued asset that developers could use for one-time payments. Session payments, which are designed for repeated or metered interactions, currently use XRP payment channels.
Wallet controls for autonomous software
OWS addresses a different part of the problem: how an agent can ask a wallet to sign or send a transaction without receiving the wallet's raw private key. RippleX says the kit can connect to an encrypted local wallet vault and apply policies before a signing request is approved. Those controls can include spending limits or other conditions set by the wallet operator. The approach is meant to reduce the amount of sensitive key material exposed to an AI application, although developers still need to assess the security of the software, policies and environment they deploy.
The release also adds xrpl-up, a command-line tool for creating a local XRP Ledger development environment, and an XRPL decentralized-exchange skill that gives agents structured instructions for common trading functions. These additions are developer tooling rather than a new consumer payment product.
What remains unfinished
The limits are material. RippleX labels the MPP software development kit as beta. Stablecoin and other issued-asset support for session payments is not yet available through XRPL payment channels; RippleX says that capability depends on proposed amendment XLS-93d. A proposal is not the same as an activated network feature, and timing remains uncertain.
The announcement also does not identify commercial customers using the new MPP connection or provide payment-volume data. Support for standards associated with Stripe and Tempo should not be read as a commercial partnership or evidence of adoption. For developers, the practical change is narrower but still notable: XRPL payment and wallet functions can now be tested within a broader set of open interfaces being developed for agent-driven commerce. Service operators evaluating the kit will still need to decide which assets they accept, how payment failures are handled and what limits apply to an automated payer. Wallet policies can reduce risk, but they do not remove the need for application-level permissions, monitoring and recovery procedures.
Sources
- The XRP Ledger AI Starter Kit v1.1: Open Standards, Not an Island
- Introducing the Machine Payments Protocol
- Ripple Adds XRP Payments to Stripe and Tempo's AI Standard in New Developer Kit
AI-generated editorial image; not a photograph of the reported event. Prepared with AI assistance and source verification.
