Reference Architecture for AI Agent Access to SFTI APIs

The white paper was developed under the leadership of our member Acrea, with input from a diverse sounding board representing SFTI’s membership. This collaborative approach demonstrates how perspectives from various domains strengthen the quality of the final results. This document provides a thorough analysis and guideline on how SFTI APIs can be improved for AI agent access.

The white paper explores how agentic AI systems and open finance are converging and what this means for the future of access to financial data and services. As AI agents evolve from simple assistants into autonomous actors capable of planning and executing complex tasks, financial institutions face a new interaction paradigm: customers increasingly engaging through personal or specialized AI agents rather than traditional digital channels.

The paper provides a strategic and technical perspective on how financial APIs can be made accessible to AI agents in a secure, standardized, and consent-based manner. It introduces emerging standards – most notably the Model Context Protocol (MCP) – and explains how they enable capability discovery, controlled execution of business functions, and bidirectional interactions between agents and service providers. Using a forward-looking pension advisory example, the paper illustrates how agentic systems could transform advisory services, operational efficiency, and client experience across the financial ecosystem.

Beyond architecture, the white paper addresses security, data protection, and operational implications of agentic access, highlighting new risks, mitigation strategies, and best practices relevant to regulated environments. It concludes with concrete recommendations for SFTI, Common API working groups, and financial institutions, outlining practical next steps to prepare for agent-based access models and to actively shape the emerging agentic financial ecosystem.

We would like to express our gratitude to the members of the sounding board for their expertise and engagement in the preparation of this white paper.

White Paper: Reference Architecture for AI Agent Access to SFTI APIs

The SFTI Reference Architecture for AI Agent Access to Open APIs core-project team consists of representatives from SFTI and Acrea.

The project is co-led by Marco Seiz, Principal Consultant at Acrea and Christof Dornbierer, Partner at Acrea.