Tuesday, February 10, 2026
The era of the Agentic Enterprise has officially arrived. CIOs and IT leaders are rapidly experimenting with AI agents across their organizations, driven by the promise of faster decision-making, automation at scale, and smarter customer and employee experiences. According toIDC, the number of actively deployed AI agents is expected to surpass one billion worldwide by 2029 a fortyfold increase from 2025.
But this explosive growth has introduced a new and growing challenge: agent sprawl. AI agents are being deployed across teams, functions, and cloud platforms, often without centralized oversight. As a result, organizations struggle to answer basic but critical questions what agents are running, what data they access, how they reason, and what outcomes they influence. Left unmanaged, this sprawl creates the risk of shadow AI, governance gaps, and operational blind spots.
To address this challenge, Salesforce has introduced new enhancements to MuleSoft Agent Fabric, designed to bring order to fragmented AI environments. These updates create a single, governed control plane that unifies AI agents, tools, and metadata across the enterprise, helping organizations move from experimentation to scalable, trusted deployment.
At the center of this update are Agent Scanners, a new capability that automatically discovers and catalogs AI agents across multiple ecosystems. Instead of relying on manual audits or spreadsheets, teams can now detect agents running on platforms such as Salesforce Agentforce, Amazon Bedrock, and Google Cloud Vertex AI in minutes. The scanners continuously monitor these environments, identifying new or updated agents and capturing what they are designed to do.
Visibility goes deeper than simple discovery. Agent Scanners extract detailed metadata, including an agent’s capabilities, the large language models powering it, and the data it is permitted to access where available. This information is normalized and mapped to Google Cloud’s agent-to-agent protocol standards, enabling consistent governance across platforms. Everything discovered is automatically synced into MuleSoft Agent Registry, creating a continuously updated catalog that replaces static, outdated documentation.
This approach removes friction for both engineers and governance teams. Instead of manually searching cloud environments, an AI engineer can see agents built for inventory, logistics, or customer service side by side—regardless of where they were created. As Brad Ringer, Enterprise & Integration Architect at AT&T, notes, MuleSoft is becoming a key accelerator for scaling long-term AI strategies in an environment where change is constant.
Beyond discovery, MuleSoft Agent Fabric also supports flexible registration for homegrown agents and Model Context Protocol servers, ensuring no AI asset is overlooked. With tools like MuleSoft Agent Visualizer, organizations gain a clear, searchable view of their entire AI footprint, making it easier to optimize performance, enforce policy, and plan for scale.
As Andrew Comstock, SVP and GM of MuleSoft at Salesforce, explains, the organizations that succeed over the next decade will be those that embrace multicloud AI innovation without sacrificing visibility and control. By unifying AI agents across platforms, MuleSoft Agent Fabric lays the foundation for enterprises to scale confidently turning agent sprawl into a strategic advantage rather than a liability.
To join our expert panel discussions, reach out to info@intentamplify.com
Recommended News
Tags:
Bell Integration expands its AI Center of Excellence in the UAE and launches NICE CXone training to help organisations modernise AI-driven customer experience operations.
HCLTech has signed a definitive agreement to acquire Singapore-based Finergic Solutions, strengthening its wealth management and core banking transformation capabilities
