AppOmni, Salt Security Unveil MCP Servers at RSA

Anthropic AI, the company behind the popular Claude family of AI models, made its Model Context Protocol (MCP) available to the industry in November, to easily and securely connect large language models (LLMs) and AI agents to external data sources and tools for efficient operations and scalability.It’s a common, open standard that any AI model can leverage to integrate with thousands of apps and data sources that also use the protocol.“Think of it as a ‘USB port’ for AI: a standardized way for any AI assistant to connect to any data source or service without requiring custom-built integrations for each connection,” Erich Hellstrom, head of prompt engineering platform provider PromptLayer, highlighted in a blog post. “Before MCP, integrating LLMs with different data sources was a significant hurdle. Each connection required bespoke code, which led to a fragmented and difficult-to-scale architecture. MCP aims to replace these ad-hoc integrations with a single, open protocol, streamlining the process and fostering a more interoperable AI ecosystem.”The protocol “represents a significant step towards a more open, interoperable, and capable AI ecosystem,” Hellstrom added.

Making a Show of IT at the Show

MCP has been widely embraced in the AI and development fields and was the subject of a number of announcements at this week’s RSA Conference 2025 in San Francisco.One of those vendors was AppOmni, which announced that its AI-powered software-as-a-service (SaaS) security tool, AskOmni, now runs as MCP server. Through this, AI agents and security platforms can more easily integrate with a range of security solutions, including network detection and response (NDR), extended detection and response (XDR), and identity and access management (IAM).Such integration capabilities make MCP servers crucial in a fast-evolving AI industry, including in the security field, where data is complex and comes in high volumes, Melissa Ruzzi, director of AI at AppOmni, told MSSP Alert.The easy integration “allows teams to easily cross over the different tools while leveraging all their AI-powered specialized benefits,” Ruzzi said. “The MCP server makes it much easier for the security tools to interconnect.”

MSSPs Get a MCP Boost

It also helps MSSPs and other channel providers, Joshua Smith, GSI partner director with the San Francisco-based company, told MSSP Alert.The standardized interface and seamless integration with security tools “facilitates a more cohesive security ecosystem,” Smith said. “By leveraging the AppOmni MCP server, MSSPs can access deep insights into SaaS identities, security postures, data exposures, and user behaviors, empowering them to conduct more thorough threat investigations and analyses, leading to faster and more targeted remediation actions.”The integration of insights reduces the manual effort needed to interpret SaaS configurations and data about activity, which streamlines operations and lets MSSPs improve their efficiencies and more easily scale the delivery of their services, he said.

MCP for API Security

AppOmni wasn’t alone in making MCP server announcements at RSA. API security specialist Salt Security rolled out its MCP server, which the Palo Alto, California company said combines its API intelligence and AI-native workflows to enhance context, insight, and control.The new tool is important with the emergence of AI agents, according to Michael Callahan, Salt Security’s chief marketing officer. They reshaping how enterprises operate but also driving three to 10 times more API traffic, which he calls a “surge of invisible risk.”“Every AI prompt, every microtask, every data fetch, it all travels through your APIs,” Callahan highlighted in a blog post. “And traditional security models, built for yesterday’s web traffic, are simply not designed for this new reality. Without a better way to manage AI-driven API interaction, organizations will find themselves vulnerable to invisible threats, hidden data leaks, and unchecked access.”With the new MCP server, organizations can better understand their API fabric, govern interactions, and respond to risks, he said.

Cloudflare’s Busy MCP Week

Cloudfare last week said it added four features to its cloud that will make it easier for organizations to build and deploy remote MCP services to its cloud, including an OAuth Provider for authorizations, McpAgent built into the Cloudflare Agents SDK to manage remote transport, and the MCP Remote adapter that lets MCP clients to work with MCP servers.In addition, its AI Playground chat interface is now a remote MCP client that allows user sto connect to remote MCP servers, with the authentication check included.This week, the San Francisco company said organizations can connect to its MCP servers from Anthropic’s Claude AI systems and from other MCP clients, including Cursor, Windsurf, and Cloudflare’s AI Playground. The service provider also is partnering with Anthropic to bring remote MCP to Claude users and showed how other companies, including Atlassian, Sentry, and Webflow, have built MCP servers on Cloudflare.