nCino (NASDAQ: NCNO) has shipped Mortgage MCP, a capability that allows lenders to connect Model Context Protocol-compatible AI agents directly to the nCino Mortgage Suite. The release uses MCP, an open-source protocol that provides a standardised interface between AI agents and external software platforms, to extend the reach of agents a lender may already be running into nCino’s own loan origination and point-of-sale environment.
The company positions the feature as a solution to a productivity bottleneck familiar to mortgage operations teams: navigating multiple screens to complete tasks that are individually simple but cumulatively time-consuming. Under the new architecture, an agent executes those tasks on the user’s behalf while remaining constrained by the lender’s existing permissions and governance settings.
Two pre-built tools at launch
Mortgage MCP ships with two pre-configured toolsets. The Admin MCP is aimed at mortgage systems administrators who manage user roles, state licences, branch hierarchies and system configurations. nCino says those tasks can now be handled through a natural-language conversation rather than by logging into the nCino console directly. The second tool, the Loan Officer MCP, gives origination staff a single conversational interface to check loan status, triage pipelines, trigger income and asset verifications, run automated underwriting system queries and draft borrower disclosures.
Casey Williams, general manager of global mortgage at nCino, said: “With Mortgage MCP, administrative workflows that used to consume hours become a five-minute conversation. Admins can connect the AI agent they already use directly into the nCino Mortgage Suite, and it only ever acts within their existing permissions, with every action logged.”
The audit-logging framework is a notable detail for compliance-conscious buyers. Every action taken through Mortgage MCP is recorded with a timestamp, a description and an outcome within nCino’s system. Configurable confirmation gates apply to high-impact operations, such as archiving a loan officer or restructuring a branch hierarchy, so that human approval is required before the agent proceeds.
Market context and regulatory read-across
The adoption of MCP as an integration standard is accelerating across enterprise software. Several major cloud and productivity vendors have begun publishing MCP servers, and the protocol is emerging as a de facto connection layer for agentic workflows in a similar way that REST APIs standardised web integrations a decade ago. nCino’s decision to build on MCP rather than a proprietary API reflects that trend and lowers the friction for lenders that are already deploying third-party AI agents from providers such as Microsoft Copilot or Anthropic Claude.
For regulated lenders, the governance architecture matters as much as the feature set. US mortgage origination is subject to oversight from the Consumer Financial Protection Bureau and a patchwork of state licensing regimes, and any agentic layer that touches loan records, borrower disclosures or officer licences will draw scrutiny from compliance and legal teams. nCino’s framing of the product around its existing permissioning and audit framework is a deliberate response to that concern, positioning the agent as a supervised actor rather than an autonomous one.
nCino reports more than 2,700 customers globally, spanning community banks, credit unions, independent mortgage banks and larger financial institutions. Mortgage MCP is available to existing customers through their nCino relationship manager.
The post nCino Adds MCP Layer to Connect AI Agents to Mortgage Suite appeared first on The Fintech Times.
