Connect Cloudmon to an AI provider to unlock two capabilities: a conversational assistant that answers questions about your live network, and in-context log analysis that explains SNMP Traps, Syslog messages, and Windows Event Log entries in plain language.
Cloudmon AI is powered by a single AI provider configuration that enables two features across the platform. Once configured, the AI assistant is available as a chat panel that you can open at any time to ask questions about your environment, and the same assistant opens automatically when you click the AI button next to a log entry in SNMP Traps, Syslog, or Windows Event Logs views.
Both features are enabled by a single AI provider configuration. Cloudmon uses your own account with the provider: bring your own API key, or connect any OpenAI-compatible endpoint. To set this up:
| Field | Description |
| Provider | The AI service to connect to: Azure OpenAI, OpenAI, Anthropic, DeepSeek, IFM, or Custom (OpenAI-compatible) for any other OpenAI-compatible endpoint. |
| API Endpoint | Required for Azure OpenAI, as the endpoint URL of your deployment, and for Custom (OpenAI-compatible). For the other providers, leave it blank to use the default endpoint, and fill it in only if you reach the provider through a proxy or gateway. |
| API Key | Required. The API key from your account with the provider. |
| Default Model | Optional. The model to use, for example gpt-4o-mini. You can select from the list or type a model name directly. Leave it blank to pick from the live list. For Custom (OpenAI-compatible), enter the model name exactly as the endpoint expects it. |
For instructions on setting up an Azure OpenAI resource and obtaining an API endpoint and API key, refer to the Azure Portal documentation for Azure OpenAI Service.
Once the AI provider is configured, the Cloudmon AI widget is accessible from the AI button in the top bar of the page. Click it to open the assistant panel. The panel shows the active AI provider and model, and a text field where you can type any question about your network.
Three preset prompts are available for common queries:
You can also type any free-form question in the input field, such as asking about a specific device, a particular alarm, or the overall health of your environment. Cloudmon AI queries your live monitoring data and returns a response directly in the panel.
The AI button appears next to individual log entries in the following views: Logs → System Events → SNMP Traps, Logs → System Events → Syslog, and Logs → System Events → Windows Event Logs. Clicking it opens the Cloudmon AI chat panel with an analysis of that entry already loaded. The analysis covers four areas:
| Section | What it contains |
| Overview | A plain-language explanation of what the log entry represents and what triggered it. |
| Impact | An assessment of the potential effect on your infrastructure, services, or users if the event is not investigated. |
| Summary | A concise statement combining the overview and impact into a single actionable description. |
| Troubleshoot and Analyse | Specific steps and areas to investigate to diagnose and resolve the event, tailored to the log type and content. |
After the initial analysis loads you can continue the conversation in the same panel, asking follow-up questions about the event or your environment.
Cloudmon AI can make mistakes. Always verify critical information directly in the monitoring views before taking action.
Beyond analyzing individual log entries, Cloudmon AI can be used as a conversational assistant. You can ask it questions in natural language and follow up within the same conversation.
Voice input: Questions can be asked by voice as well as by typing, so the assistant can be queried hands-free and responds in the same conversation.
Chat history: Conversations are retained as chat history, so earlier questions and the assistant's responses stay available to review and continue later, rather than being lost when you navigate away.