AI agents in Business Applications

Overview

Turbo360 brings AI-powered intelligence directly into your Business Applications workflow through AI agents, purpose-built assistants that analyze your Azure resources in real time and surface insights you can act on immediately.

Rather than manually inspecting configurations or cross-referencing Azure documentation, you can simply pick an agent, hit Generate, and receive a structured analysis of your selected resource, streamed live via Azure OpenAI, always reflecting the resource's current state.

Note:

AI-generated content: The analysis produced by AI agents is generated by Azure OpenAI, based on your resource's current state at the time you run the agent. Review the output before applying recommendations to your environment.

Business value

AI agents eliminate the manual effort of resource audits, security reviews, and monitoring configuration. Instead of switching between Turbo360 and Azure documentation, you get actionable, context-aware analysis directly within your Business Application, reducing time to insight and helping your team respond faster to configuration gaps, security risks, and performance issues.

For Logic App workloads, the cost review agent adds financial visibility, surfacing spend trends and savings opportunities without leaving the Business Applications context.

How it works

Turbo360 offers four specialized AI agents for Business Applications. Each agent targets a specific operational concern and generates its analysis on demand, scoped to the resource you select.

AI resource overview: get an instant, plain-language summary of what your resource does, how healthy it is, and what you should look at next. This agent is particularly useful when onboarding to an unfamiliar environment or doing a quick resource audit.
AI security checklist: assess your resource against Azure security best practices automatically. This agent identifies hardening opportunities, organizes findings by security pillars, and recommends targeted fixes so you can close gaps without manually running through checklists.
AI metrics advisor: skip the guesswork on Azure Monitor configuration. This agent automatically recommends the right metrics for your resource based on Azure SRE best practices, complete with pre-filled warning and error thresholds, aggregation types, and priority tags (Critical, Important, or Optional) so you can review, fine-tune, and apply your monitoring setup in one go.
Logic App cost review: get a structural cost breakdown of your Logic App workflow with concrete savings recommendations covering monthly spend trends, bill spike risks, and prioritized next steps. Available exclusively for Logic App Consumption and Standard hosting plans.

In addition to the standalone agents above, Business Applications surfaces two inline AI features directly within the monitoring rules configuration: Explain metric with AI and Explain rule with AI. Unlike the standalone AI agents, these are not launched from the AI agents panel. They appear as a sparkle icon (✦) next to individual rule and metric names in the monitoring configuration UI and open a contextual explanation panel scoped to that specific rule or metric.

Architecture

When you run an AI agent in Business Applications, the request passes through three components inside the Turbo360 SaaS platform:

Angular SPA is the browser-based interface that captures your prompt input and renders the streamed analysis as it arrives. Responses appear progressively so you can start reading immediately while the model is still generating.

Prompt Builder & Orchestrator is a server-side layer that assembles the full prompt before sending it to Azure OpenAI. It gathers the relevant resource context (configuration data, Azure Monitor metrics, security signals, Logic App workflow structure), applies guardrails to scope and sanitize the request, and injects telemetry so the model analyzes your resource's actual current state rather than making generic assessments.

Azure OpenAI receives the assembled prompt and generates the response using GPT reasoning models and domain-specific agents. The response is streamed back to the Angular SPA using Server-Sent Events (SSE), delivering output token by token directly to your browser.

Turbo360 AI Agents architectural diagram.png

This architecture applies to all AI agents in Business Applications on SaaS deployments. On Private Hosting, the same flow applies, but the Azure OpenAI component is replaced by the Azure OpenAI resource you configured under Settings > OpenAI connection. See OpenAI setup in Turbo360 for the Private Hosting setup procedure.

AI agents account-wide toggle

The Account owner can enable or disable AI agents for the entire organization from a single toggle. When disabled, AI agents are hidden across the app for every user, regardless of their individual role permissions.

The toggle location depends on your deployment model:

SaaS: navigate to Settings > General settings. The AI agents toggle is in the right panel.
Private hosting: navigate to Settings > OpenAI connection. The AI agents toggle appears on the same page as your connection configuration.

Both the account-wide toggle and the per-role AI agents permission must be enabled for a user to access AI agents.

Data and AI transparency

This section describes what data Business Applications sends to Azure OpenAI when AI agents features are used, and how that data is handled.

What data is sent

When you run an AI Agent, Business Applications assembles a prompt from your resource's current data and sends it to Azure OpenAI to generate the response. The data included depends on the agent, but covers one or more of the following categories:

Resource metadata: resource name, type, kind, location, resource group, and subscription name
Resource configuration data: resource-level settings and properties retrieved from Azure Resource Manager, used by the AI resource overview and AI security checklist agents
Security signal data: identity configuration, TLS version, network access settings, and governance tag status, assessed against Azure security best practices for the AI security checklist
Azure Monitor metrics: recent performance data including aggregates and thresholds for the resource type, used by the AI metrics advisor and Explain metric with AI
Monitoring configuration: configured warning and error thresholds for the specific metric being explained, used by the Explain metric with AI feature
Logic App workflow structure: trigger count and types, action count and types, and connector identifiers. Workflow step configuration, HTTP URLs, message bodies, and connection credentials are not included. Used by the Logic App cost review agent only.
Cost and usage data: monthly spend figures and cost trends from Azure Cost Management, used by the Logic App cost review agent

Azure resource tags are included for agents that analyze resource overview and security. Tags help the model understand resource context and ownership.

What data is not sent

Your Azure service principal credentials are used server-side to retrieve resource data from Azure and are never included in the prompt sent to Azure OpenAI.
Logic App workflow step configuration, including HTTP URLs, message bodies, parameter values, and connection credentials, is never sent. Only aggregate counts, action types, and connector identifiers are included.
For AI metrics advisor, your currently configured warning and error threshold values are deliberately excluded from the prompt to keep recommendations independent of your existing setup.

Response storage and logging

AI agents responses are streamed directly to your browser and are not stored in any Turbo360 database or server cache. Raw prompts sent to Azure OpenAI are not logged.

SaaS and Private Hosting

The data sent to Azure OpenAI is the same regardless of deployment type. The difference is which Azure OpenAI resource receives it:

SaaS: your resource data is sent to a Turbo360-managed Azure OpenAI connection. Turbo360 manages this connection on your behalf; no configuration is required from you.
Private Hosting: your resource data is sent to the Azure OpenAI resource you configured under Settings. The data stays within the connection you own and control.

Permissions

AI agents in Business Applications are accessible to the following users:

Account Owner: has full access to all features in Business Applications, including AI agents, by default.
Owner: has complete access to Business Applications, including AI agents, by default.
Contributor: has access to manage everything at the scope at which it is assigned, including AI agents, by default.
Custom Role: users with a custom role must have the AI agents feature permission explicitly set to Enabled.

Steps

Use the following steps to access and launch AI agents within Business Applications. Navigate to your Business Application to get started.

AI resource overview and AI security checklist

The AI resource overview and AI security checklist agents give you on-demand resource summaries and security assessments. They are accessible from three levels within a Business Application.

Resources level: from the overall Resources section of a Business Application, click the AI agents icon next to any resource in the list to launch the AI agents panel.
Resource type level: navigate into any specific resource type (e.g. API Apps, Logic Apps, Azure Function Apps), and click the AI agents icon next to the resource within that section.
Resource level: open an individual resource and click the AI agents button in the top action bar to launch the panel directly from within the resource.

AI metrics advisor

The AI metrics advisor agent generates resource-type-specific metric recommendations you can apply directly to a monitoring configuration or profile. It is accessible from two places within a Business Application.

Resource monitoring tab: open any resource within a Business Application, navigate to the Monitoring tab, and click AI metrics advisor in the Metrics section to get recommendations for that specific resource.

Monitoring profile (Create/Edit): while creating or editing a monitoring profile, select a resource type and click AI metrics advisor in the Metrics section to apply AI-recommended metric thresholds directly to the profile.

Explain with AI

Explain with AI is an inline AI feature available directly within the monitoring rules configuration. Unlike the standalone AI agents, it is scoped to a single rule or metric and opens as a panel without leaving the configuration view. It is available in two variants depending on the rule type.

Explain metric with AI is available for every metric rule, including metrics such as Instance Count and all other Azure Monitor metrics for the resource. Click the sparkle icon (✦) next to a metric name in the Metrics section, available both within an individual resource's Monitoring tab and in the Add monitoring profile or Edit Resource rules panel. The explanation panel includes:

Metric overview: what the metric measures, its unit, and its aggregation behavior.
Why monitor it on this resource: a context-specific explanation based on the resource's current state and region.
Healthy vs unhealthy range: a visual range indicator showing the suggested warning and error thresholds alongside the current observed baseline.
Likely causes of a violation: the most common reasons this metric breaches its threshold, with operational context.
Recommended response on alert: suggested actions to take when this rule is violated.
Suggested thresholds: AI-recommended warning and error threshold values derived from typical values for this resource type.
Recommended setup: the recommended aggregation type for the metric with a rationale.
Monitor these together: related metrics worth pairing with this rule for a fuller picture of the resource's health.
Common setup pitfalls: aggregation and threshold mistakes to avoid for this specific metric.

Explain rule with AI is available for Availability Status rules and resource State rules. Click the sparkle icon (✦) next to a rule name in the Availability status or Status section of the monitoring configuration. The explanation panel includes:

What this rule checks: a plain-language description of what the rule monitors and what a violation means.
Why monitor this rule: a context-specific explanation of why the rule matters for the selected resource type.
Healthy vs unhealthy state: the expected healthy states and what it means when the violation state is reached.
Likely causes of a violation: the most common reasons this rule fires, with operational context.
Recommended response on alert: suggested actions to take when this rule is violated.
Expected state: the configured condition and violation state for this rule.
Recommended setup: guidance on the correct condition setting and common configuration mistakes.
Monitor these together: related rules and metrics worth pairing with this rule for fuller coverage.
Common setup pitfalls: configuration mistakes that lead to noisy or missed alerts for this rule type.

Troubleshooting

AI agents option is not visible for a user
Cause: The user has a Custom Role that does not have the AI agents feature permission enabled.
Fix: Go to User management > Roles, edit the relevant Custom Role, and set the AI agents permission to Enabled under Specify feature permissions.
The sparkle icon (✦) for Explain with AI does not appear next to a rule or metric
Cause: Explain with AI is only available for metric rules (Explain metric with AI) and Availability Status or State rules (Explain rule with AI). It does not appear for Property or non-supported rule types.
Fix: Confirm the rule type is a metric, Availability Status, or State rule. If the icon is expected but missing, check that your OpenAI connection is active (Private Hosting only).
AI agents panel opens but generates no output on Private Hosting
Cause: The Azure OpenAI connection has not been configured, or the configured endpoint or key is invalid.
Fix: Go to Settings and verify your Azure OpenAI connection details. See OpenAI setup in Turbo360 for the full setup procedure.
Logic App cost review is not available for a Logic App resource
Cause: The Logic App is not using a Consumption or Standard hosting plan. Logic App cost review is only available for these two hosting plans.
Fix: Confirm the Logic App's hosting plan. If it uses a different plan type, the Logic App cost review agent is not supported for that resource.
AI metrics advisor recommendations appear but cannot be applied
Cause: The user does not have Manage permission on the Monitoring feature for that resource.
Fix: Confirm the user has at least Manage access to the resource's Monitoring feature permission before attempting to apply recommendations.
AI agents are not visible to any user despite role permissions being correct
Cause: The account-wide AI agents toggle has been disabled by the Account owner.
Fix: Navigate to Settings > General settings (SaaS) or Settings > OpenAI connection (Private hosting) and confirm the AI agents toggle is enabled.

FAQs

Do I need to configure an OpenAI connection to use AI agents on SaaS?
No. SaaS deployments use Turbo360's in-house Azure OpenAI connection and require no additional setup. AI agents is available immediately. For Private Hosting deployments, you must configure your own Azure OpenAI resource under Settings before AI agents can generate output. See OpenAI setup in Turbo360 for steps.
Which roles can access AI agents by default?
Account Owner, Owner, and Contributor have full access by default. Users with a Custom Role must have the AI agents feature permission explicitly set to Enabled in their role configuration.
What is the difference between AI agents and Explain with AI?
AI agents are standalone features launched from the AI agents panel or button on a resource. Each covers a broad operational concern such as security, cost, or monitoring setup. Explain with AI (Explain metric with AI and Explain rule with AI) is an inline feature that opens directly from the monitoring configuration panel next to a specific metric or rule, providing a targeted explanation scoped to that single item.
Does Explain with AI work inside monitoring profiles?
Yes. The sparkle icon (✦) is available next to metric names when creating or editing a monitoring profile, so you can get AI-generated guidance on any metric rule without leaving the profile editor.
Is my resource data stored after an AI Agent or Explain with AI runs?
No. Responses are streamed directly to your browser and are not stored in any Turbo360 database or server cache. Raw prompts sent to Azure OpenAI are not logged.
Can the Account owner disable AI agents for all users?
Yes. The Account owner can disable AI agents organization-wide from Settings > General settings (SaaS) or Settings > OpenAI connection (Private hosting). When disabled, AI agents are hidden for all users regardless of their role permissions.