---
title: "AI agent integration with MCP"
summary: "Connect an AI assistant to the OpenSanctions API with the Model Context Protocol."
url: "https://www.opensanctions.org/docs/api/mcp/"
date_updated: "2026-11-13"
---

The OpenSanctions MCP server exposes the [hosted API](/api/) to LLM agents over the [Model Context Protocol](https://modelcontextprotocol.io/). An assistant connected to it can screen people, companies, and other entities against sanctions, PEP, and watchlist data, walk relationship graphs, and audit where a claim came from.

## Install

* **Claude Desktop:** [add OpenSanctions as a connector](https://claude.ai/customize/connectors?modal=add-custom-connector&connectorName=OpenSanctions&connectorUrl=https%3A%2F%2Fmcp.opensanctions.org%2Fmcp).
* **ChatGPT:** [open the connector settings](https://chatgpt.com/#settings/Connectors) and create a connector with the URL `https://mcp.opensanctions.org/mcp`.
* **Other applications:** add `https://mcp.opensanctions.org/mcp` as a remote MCP server.
## More information

The [`yente-client` documentation](https://yenteclient.followthemoney.tech/mcp/) describes the tools that the server provides and how to run it with an on-premise deployment.
