AppOmni Adds MCP Server to Platform for Protecting SaaS Applications
AppOmni at the 2025 RSA Conference today added a Model Context Protocol (MCP) server to its platform for protecting software-as-a-service (SaaS) applications.
Originally developed by Anthropic, MCP is emerging as a de facto standard for integrating artificial intelligence (AI) agents and models.
Melissa Ruzzi, director of AI for AppOmni, said it’s now only a matter of time before AI agents will be pervasively employed. MCP will enable AppOmni to more easily extend the integration capabilities that AppOmni already provides to AI agents, including generative AI agents that AppOmni has developed to automate cybersecurity workflows, and tools and platforms provided by other providers of cybersecurity tools and platforms, she added.
In general, AI agents are about to significantly reduce the current level of complexity organizations experience when integrating SaaS applications. In fact, AI agents in time might ultimately eliminate the need to master a different set of user interfaces for each application.
Those same capabilities will also soon be applied to cybersecurity workflows spanning multiple tools and platforms using MCP servers, said Ruzzi.
AI agents, however, also present cybercriminals with an additional set of rich targets that enable them to compromise workflows by, for example, rerouting requests. Previously, AppOmni developed a security posture management platform for SaaS applications that identifies threats and misconfigurations. That capability via an MCP server can now be applied to AI agents in a way that also makes it simpler to integrate cybersecurity tools.
This capability is arriving at a time when cybersecurity teams have been debating the merits of adopting integrated platforms versus continuing to rely on a range of best-of-breed tooling. One of the key arguments for adopting an integrated platform is to reduce the need for cybersecurity teams to master multiple individual tools. However, with the rise of AI agents the need to master different user interfaces will become less of an issue.
Cybersecurity teams may, of course, be able to reduce the total cost of cybersecurity by relying more on integrated platforms but the cost of acquiring those platforms can be significant.
It’s not entirely clear how AI agents will transform the way cybersecurity is managed, but it appears there will soon be much less need to master programming languages to automate workflows. Instead, AI agents connected via MCP servers can be more simply assigned a set of tasks that cybersecurity teams will be able to more easily orchestrate.
That’s critical in an era now where cybersecurity teams need to be able to effectively respond to cyberattacks in minutes. The longer it takes to investigate those threats, the more damage there is likely to have been inflicted.
It’s not clear to what degree cybersecurity teams are embracing AI, but it’s now only a matter of time before these capabilities are embedded into almost every tool they employ. Each cybersecurity team will need to determine to what degree to trust the output created by AI models and agents, but much of the toil that conspires to burn these teams out should be sharply reduced in the month ahead.
The challenge now is determining which tasks can be reliably assigned to AI agents that are supervised by cybersecurity professionals that should now have the time needed to investigate more advanced threats.