Network operations teams pay a tax every time their footprint grows. More sites, more gateways, more overlay networks, and suddenly the data you need to answer any one question is scattered across half a dozen places. Troubleshooting in the heat of the moment is hamstrung by the need to gather information from multiple places. Even within a single system like Trustgrid, answering a question can require jumping between multiple screens to piece together what is happening.
When a site goes sideways, the network is rarely the bottleneck. The bottleneck is the half hour a tired engineer spends pulling data out of five places before they can form a hypothesis.
Reporting is the same story. The desired answer almost always lives behind a custom query or a CSV export that somebody has to massage by hand. None of it is difficult, it’s just slow and easy to get wrong.
A new class of AI-enabled tooling is changing that. Instead of jumping between consoles, you describe what you need in plain English and the data (or the diagnosis, or the report) comes back without you opening another tab. The AI assistant is rapidly emerging as the “single pane of glass” through which network operations teams manage their infrastructure.
Today we are announcing the Trustgrid MCP server. It lets your AI assistant of choice talk directly to your Trustgrid environment. Using the new MCP capabilities, you can perform ad-hoc reporting, walk a virtual network topology to track down a misconfigured route, or run live diagnostics on a remote node while you are on the phone with the customer at that site.
AI-Enabling Your Network Operations Team
The Model Context Protocol (MCP) is an open standard that gives AI agents a uniform way to talk to external systems. An MCP server exposes a set of tools (read APIs, on-node diagnostics, documentation search) and any MCP-capable AI client can connect to it, discover what is available, and invoke those tools on your behalf.
The agent handles the work of breaking down your request, picking the right tools, and folding the results into an answer. Anthropic originally developed MCP and every major model provider now supports it, so whichever AI client your team prefers will already know how to talk to the Trustgrid MCP server.
Custom Ad-hoc Reports On Demand
Instant custom reports containing deployment status, site traffic statistics or incident history can now be generated instantly. Trustgrid MCP allows the user to query and filter by whichever metrics make sense, and present them in the desired output format.
Examples
- Show me all gateway nodes that are running an old software version
- Show all unresolved alerts, including site, alert type and the time since creation
Rapid Real-time Incident Troubleshooting
The MCP server can be used to perform rapid data gathering for incident investigation. It can pull live node runtime status, run remote endpoint reachability checks, sniff local network interfaces, and simulate packet paths at the sites where the problem is happening. This allows NetOps to drive a whole incident investigation from a prompt or runbook written in plain English.
Examples
- Investigate why there appears be excessive dataplane latency between site X and site Y
- Is the endpoint 1.2.3.4:443 reachable at site X?
- Are there any half-open TCP connections at site X?
Network Topology Traversal
The operator can summarize or dig into node configurations, virtual network topologies and site relationships. This allows the operator to gather the network context they need in seconds rather than minutes of clicking through tabs trying to relate the relevant components needed to get the complete picture.
Examples
- Summarize configuration of virtual network X, including all network configuration changes from the past week.
- Show me the life of an incoming tcp packet 1.2.3.4 -> 5.6.7.8:443 on the LAN at site X
All of these use cases are performed with fine-grained permissions that you actually control. The MCP server honors only the permissions assigned to the user in the Trustgrid portal, and you can scope down further at the OAuth consent step so the agent only sees what the task requires.
Optional Codemode
Codemode is a second way of talking to the Trustgrid MCP server, and for some workloads it is the better option. When using Trustgrid MCP codemode, a multi-step query that would otherwise mean six tool calls and a lot of intermediate data trickling through the model usually finishes in a single turn and burns a fraction of the tokens.
Codemode is a good fit for anything that involves filtering or aggregating across Trustgrid resources such as flow logs across sites, alerts grouped by region, route tables walked across clusters.
Trustgrid MCP Changes the NetOps Experience
The Trustgrid MCP server fundamentally changes the bottleneck in network operations from the human engineer gathering scattered data to an AI-enabled assistant providing instant answers and actions.
By implementing the open Model Context Protocol, Trustgrid transforms troubleshooting and reporting into simple, plain-English prompts, delivering superpowers to NetOps teams. Whether generating instant ad-hoc reports, driving rapid real-time incident investigations, or traversing complex topologies, the MCP server eliminates manual, time-consuming steps while honoring fine-grained user permissions.
Trustgrid MCP is the single pane of glass your team needs to manage your infrastructure efficiently and accurately.
Learn more about the Trustgrid MCP server at https://docs.trustgrid.io/docs/mcp/
Chief Technology Officer
Steven Stites is the CTO and Co-Founder of Trustgrid, where he leads the vision and engineering teams behind the company’s innovative platform for secure networking and edge computing solutions. With over 20 years of expertise in network security, distributed computing, and cloud infrastructure, Steven brings deep industry experience to establishing Trustgrid as a trusted provider for secure, scalable application connectivity across FinTech, HealthTech, SaaS, and enterprise environments.
Leadership at Trustgrid
As CTO and Co-Founder, Steven drives the technical strategy, product development, and architectural direction at Trustgrid. He focuses on creating solutions that bridge modern hybrid ecosystems, empowering SaaS and cloud application providers to connect securely to on-premise resources with maximum reliability and performance. Steven’s guidance is central to Trustgrid’s integration of SD-WAN, Zero Trust Network Access (ZTNA), and edge computing into a unified platform, simplifying deployment, elevating data security, and supporting enterprise-grade operational scale .
Professional background
Before founding Trustgrid in 2017, Steven held senior technical leadership roles at Cisco, where he served as Senior Technical Leader for IoT Cloud and Cloud Web Security. At Cisco, he architected and led customer engagement for major SaaS security products, designing enterprise-scale networking and security solutions and overseeing technical vetting for large-scale technology acquisitions. Earlier in his career, Steven spent over a decade at IBM as a technical lead, driving development for network monitoring and distributed application performance products, and began as a software engineer researching sonar and signal processing at Applied Research Labs. He holds a bachelor’s degree in Electrical and Electronics Engineering from The University of Texas at Austin .
Building the Future of Connectivity
Steven’s vision at Trustgrid centers on advancing secure, cloud-like connectivity across modern digital environments, ensuring frictionless integration between public cloud, data center, and on-premise resources. His background in high-performance network design and distributed security shapes Trustgrid’s commitment to eliminating complexity in deploying, monitoring, and supporting thousands of application connections. He is also an inventor, with patents for secure network technologies and is recognized as a strategic leader with a rare blend of deep technical expertise and business insight .
About Steven Stites
Steven is a passionate technology executive and product architect based in Austin, Texas. His approach emphasizes pragmatic problem-solving, strong team leadership, and client advocacy, helping organizations leverage networking and security innovations to enable secure, scalable applications. He is highly regarded for his ability to clarify complex technical challenges, mentor teams, and deliver solutions that balance technical excellence with cost efficiency. Steven is deeply interested in machine learning, cloud security, and agile product development.
Connect with Steven
https://www.linkedin.com/in/srstites/
Or
Contact him at www.trustgrid.io