Salt Security Launches Salt MCP Finder Technology
New functionality uncovers invisible MCP servers and consolidates discovery right into a unified stock, giving safety groups the visibility required to handle threat within the quickly increasing Agentic AI Action Layer.
Salt Security, the chief in AI agent and API safety, right this moment introduced Salt MCP Finder know-how, the trade’s first devoted discovery engine for Model Context Protocol (MCP) servers, the fast-proliferating infrastructure powering agentic AI. MCP Finder supplies organizations with a whole, authoritative view of their MCP footprint at a second when MCP servers are being deployed quickly, typically with out IT or safety consciousness.
The invisible spine of agentic AI and the most recent blind spot in cybersecurity
As enterprises speed up the adoption of agentic AI, MCP servers have emerged because the common API dealer that lets AI brokers take motion: retrieving information, triggering instruments, executing workflows, and interfacing with inner programs. But this new energy comes with a brand new downside: MCP servers are being deployed in every single place, by anybody, with nearly no guardrails.
- Developers spin them up for prototyping.
- Business items deploy them to attach brokers to SaaS instruments.
- Vendors and contractors introduce them throughout integration tasks.
- Open-source MCP servers get dropped into repos and shipped to manufacturing.
- Internal enterprise groups are deploying MCPs alongside new inner APIs to help shadow agentic workflows, outdoors the visibility of IT or safety.
This wave of adoption sits atop fractured inner API governance in most enterprises, compounding threat. Once deployed, MCP servers develop into simply accessible, enabling brokers to attach and execute workflows with minimal oversight. This turns into a significant supply of operational publicity.
The result’s a quickly rising API material of AI-connected infrastructure that’s largely invisible to central safety groups. Organizations have no idea:
- How many MCP servers are deployed throughout the enterprise
- Who owns or controls every server
- What APIs and information every server exposes
- What actions brokers can carry out by accessible MCP instruments
- Whether company safety requirements are adopted and primary safety controls like authentication, authorization, and logging are in place
Recent trade observations present why this visibility disaster issues. One research confirmed that solely ten months after the launch of the MCP, there have been over 16,000 MCP servers deployed throughout Fortune 500 firms. Another confirmed that in a scan of 1,000 MCP servers, 33% had essential vulnerability and the typical MCP server had greater than 5. MCP is shortly changing into one of many largest sources of “Shadow AI” as organizations scale their agentic workloads.
According to Gartner® “Most tech suppliers stay unprepared for the surge in agent-driven API utilization. Gartner predicts that by 2028, 80% of organizations will see AI brokers eat nearly all of their APIs, relatively than human builders.”
Gartner additional acknowledged, “As agentic AI transforms enterprise programs, tech CEOs who perceive and implement MCP would drive progress, guarantee accountable deployment and safe a aggressive edge within the evolving AI panorama. Ignoring MCP dangers falling behind as composability and interoperability develop into essential differentiators. Tech CEOs should prioritize MCP to steer within the period of agentic AI. MCP is foundational for safe, environment friendly collaboration amongst autonomous brokers, instantly addressing belief, safety, and price challenges.”*
A prerequisite for governing the AI Action Layer
Salt’s MCP Finder know-how solves the foundational problem: you can not monitor, safe, or govern AI brokers till you already know what assault surfaces exist. MCP servers are a key part of that floor.
“You can’t safe what you’ll be able to’t see,” stated Nick Rago, VP of Product Strategy at Salt Security. “Every MCP server is a possible motion level for an autonomous agent. Our MCP Finder know-how provides CISOs the one supply of fact they should lastly reply crucial query in agentic AI: What can my AI brokers do inside my enterprise?“
Salt’s MCP Finder know-how delivers full, automated MCP stock by three discovery layers
Salt’s MCP Finder know-how uniquely consolidates MCP discovery throughout three programs to construct a unified, authoritative registry:
- External Discovery – Salt Surface
Identifies MCP servers uncovered to the general public web, together with misconfigured, deserted, and unknown deployments. - Code Discovery – GitHub Connect
Using Salt’s not too long ago introduced GitHub Connect functionality, MCP Finder inspects non-public repositories to uncover MCP-related APIs, definitions, shadow integrations, and blueprint recordsdata earlier than they’re deployed. - Runtime Discovery – Agentic AI Behavior Mapping
Analyzes actual visitors from brokers to watch which MCP servers are in use, what instruments they invoke, and the way information flows by them.
Together, these sources give organizations the one supply of fact required to visualise threat, implement posture governance, and apply AI security insurance policies that stretch past the mannequin into the precise motion layer.
Salt’s MCP Finder know-how is out there instantly as a core functionality throughout the Salt Illuminate
platform.
*Source: Gartner Research, Protect Your Customers: Next-Level Agentic AI With Model Context Protocol, By Adrian Lee, Marissa Schmidt, November 2025.
GARTNER is a registered trademark and repair mark of Gartner, Inc. and/or its associates within the U.S. and internationally and is used herein with permission. All rights reserved.
The submit Salt Security Launches Salt MCP Finder Technology first appeared on AI-Tech Park.