The new integration, through Claude and Compliance API, provides the security team with a real-time read-only list of servers in MCP of their Claude organization.
Palo Alto, California, September 29, 2026 / PRNewswire / -- Proxy-based security and API security company Salt Security today launches Salt Claude Connect. This provides security teams with continuous, read-only visibility to monitor the context protocols ( MCP ) of servers connected to their Claude Enterprise organization's models.
This integration is used in conjunction with Compliance API of Claude. It allows the security team to display in the Salt list the MCP servers managed by the organization, including when each server was added, updated, or removed, as well as its current status.
As enterprises adopt AI, the security and IT teams face a fundamental issue: what content is connected to our AI, and what can it access? The MCP servers have expanded the reach of the AI environment, and in large organizations, the rate at which these servers increase may exceed the tracking capabilities of the security teams.
The scale of related risks can now be quantified. Research by Trend Micro has found that in just a few months, the number of MCP servers exposed to the public internet without authentication or encryption protection has nearly tripled, reaching 1,467. Most of these servers are hosted on major cloud service providers, and the data behind them can be directly accessed. The company stated that without the latest inventory, the team would not be able to answer basic questions regarding the scope of their AI attacks, manage the connected content, or conduct investigations as changes occur.
Salt Claude Connect Features Provided
Salt Claude Connect reads the organization's activity stream through Compliance API of Claude, and automatically displays in Salt AI and inventory the MCP servers managed by the organization. The security team can obtain a real-time, always-up-to-date overview of which MCP servers are connected to their Claude organization, when they were added or removed, and their current status.
The discovered servers will appear in Salt Agentic and inventory, and are marked according to their source; each server will display its name, status, as well as the time of the most recent lifecycle event. As connectors are added, updated, or removed, this list will continue to be updated.
Designed according to the principles of read-only and minimal permissions.
This integration is built around the principle of least privilege access. It only requires a read-only permission scope and is intended to read only the lifecycle activities of the MCP server. It does not send prompts to Claude, nor does it read chat content, file contents, or project data. The only information stored in Salt are the lifecycle events of the MCP server, and this integration does not modify the Claude organization in any way.
Co-founder and CEO Roey Eliyahu stated: "The MCP server is one of the fastest-growing components of an enterprise's attack surface, and for most security teams, it has always been an irreparable blind spot as to what exactly is connected to their AI environment. Salt Claude Connect has turned this blind spot into a real-time inventory. Security teams can now see the MCP servers connected to their Claude Enterprise organizations, track them as they are added, removed, or modified, and ultimately manage their AI environment in the same rigorous manner as they manage other environments, without touching any prompts or content that runs within it."
Availability
Salt Claude Connect is now available to Salt Security customers who own the Claude Enterprise organization. Setting it up requires a Compliance Access Key with read-only permissions, as well as the ID organization, and can be configured through a few steps via the Salt console. For complete details, please refer to salt.security.
About Salt Security
Salt Security indicates that the company is a leader in the fields of proxy-based security and API security, dedicated to protecting the world's most innovative enterprises from AI proxies and API attacks. The company states that AI is shifting from "question-answering chatbots" to "task-executing proxies," while Salt provides security guarantees for the entire proxy-based process, covering everything from AI code generation to runtime, spanning models, MCP servers, tools, and downstream API.
Salt Security was established in 2016, with investors including Sequoia Capital, S Capital, Tenaya Capital, Salesforce Ventures, Advent International and other leading investors. For more information, please visit salt.security.
Contact person: Dr . Karl Bateson - [ email protected ]
