Sign up
Subscribe
Home / news / Sumsub adds MCP integration so AI agents can build compliance setups from AML policies
news

Sumsub adds MCP integration so AI agents can build compliance setups from AML policies

Sumsub adds MCP integration so AI agents can build compliance setups from AML policies

Sumsub has launched a Model Context Protocol (MCP) integration and a new set of AI agent skills that let AI tools such as Claude and ChatGPT work not only with day-to-day compliance tasks, but with the platform’s configuration layer itself. For high-risk operators, that means an AML policy document can be turned into a live verification setup without a human team hand-building every rule, workflow, and questionnaire.

  1. Sumsub says it is the first identity verification and compliance platform to give AI agents access to the full setup layer of the product, not just operational functions. In practice, that means an AI agent can take a real compliance policy document and generate a fully configured Sumsub environment inside the customer dashboard.
  2. The company says this can translate a complex, multi-page PDF with country-specific risk brackets, weighted scoring tables, and conditional logic into live platform settings, including verification levels, risk questionnaires, and onboarding workflows. A setup that previously could take days can now be completed in minutes.
  3. The release targets three main use cases for compliance teams: policy-to-configuration, where teams upload AML policy or regulatory requirements and ask an AI agent to configure the environment; faster technical integration, where agents handle the code needed to embed verification into an onboarding flow; and day-to-day compliance work, including applicant review, analytics, verification links, and responses to regulatory changes.
  4. Andrew Novoselsky, Chief Product Officer at Sumsub, said the company is connecting “an AI agent directly to the configuration layer of the platform,” allowing a team to hand over its AML policy and have the full environment built automatically. He framed that as “a fundamentally different category of capability” from what has been available in the space.
  5. The integration is model-agnostic and is designed to work with any leading AI agent. Sumsub has also published an open-source set of agent skills on GitHub, installable with a single terminal command. The MCP integration sits alongside Summy, Sumsub’s AI Copilot for compliance and fraud teams inside the platform.

Weekly high-risk digest

Regulation, sanctions and payment news across your verticals — once a week, free.

Please check your inbox and click the link to confirm your subscription.

Please enter a valid email address!