NetSuite AI Connector Service FAQ

What can I do with the NetSuite AI Connector Service

With the NetSuite AI Connector Service you can use your AI client to query data using natural language. You can:

• Create, read, update records

• Run reports.

• Run saved searches.

• Execute SuiteQL queries (SuiteQL tools support read-only queries only).

By default, the MCP Standard Tools SuiteApp comes with tools for working with records, reports, saved searches, and SuiteQL queries. The MCP Standard Tools SuiteApp is compatible with AI clients that:

• Support remote MCP.

• Follow the Protocol version 2025-06-18.

• Support streamable HTTP.

• Support OAuth 2.0 Authorization Code Grant flow with Proof Key for Code Exchange (PKCE).

For more information and details, see MCP Standard Tools SuiteApp

If you want, you can build your own tools. For more information, see Creating Custom Tools for the NetSuite AI Connector Service. MCP also allows orchestration with other business applications within your AI client's environment.

Do I need custom tools?

Not always. You should use the MCP Standard Tools SuiteApp, when the task is common or general.

Custom tools are better suited if you need:

• Specialized workflows

• Faster or single-step operations

• Custom automation

How do I create custom tools?

To create custom tools you need:

• SuiteScript 2.1 custom tool script

• JSON schema definition

• toolset SDF object

The key requirements for the custom script are:

• The script header must include @NScriptType CustomTool

• Asynchronous entry points

• Supported modules only

For more information, see Creating Custom Tools for the NetSuite AI Connector Service.

How do I manage custom tools?

To manage custom tools, go to Customization> Scripting > Custom Tools. There you have the options to:

• View tools

• Delete toolsets from account customization projects (ACPs) directly from the page

• Remove toolsets from SuiteApps by uninstalling the SuiteApp

For more information, see Managing Custom Tools in NetSuite.

How accurate are the results?

The MCP Standard Tools SuiteApp improves accuracy, because it retrieves data directly from NetSuite reports and saved searches. However, AI clients may still produce incorrect or "hallucinated" results, so it is best practice to always verify the results against NetSuite data.

How should I write prompts for better results?

Use clear and structured prompts:

• Define the objective

• Provide context

• Include:

  • Date ranges

  • Filters

  • Subsidiaries

A prompt with a structure like below usually returns more accurate responses:

• Instruction

• Role

• Context

• Input data

• Output format

For more information, see Prompting Guidelines.

How does authentication and access work?

OAuth 2.0 authentication leverages NetSuite roles and permissions. Your NetSuite account administrator authorizes specific roles to have access to the MCP server and MCP tools to be installed. Your administrator can review this access and make changes to it at any time.

Does AI have access to all my data?

Queries executed through the connector respect your NetSuite role's permissions, so tools can only access data you're allowed to see when logged in with this role.

Additionally, the AI client can perform only tasks defined in the tools. The MCP Standard Tools SuiteApp offers restricted tool capabilities.

Tools can't run as administrators, can't call external APIs and can't run elevated scripts. They run with user role permissions only.

How do I keep it secure?

You should review your organization's security guidance and the AI client's policies before using the NetSuite AI Connector Service. For detailed information about risks, controls, and mitigation strategies, see Associated Risks, Controls, and Mitigation Strategies.

When authorized, data sent to the third-party LLM is governed by the third-party LLM's privacy policy on data handling. You can choose the tools and the data that will be processed by the third-party LLM. Data in the NetSuite Service will be handled as per Oracle Terms of Service.

Generally, it is best practice to:

• Use trusted AI clients.

• Limit the permissions to only those really needed.

• Avoid using any high-privilege roles.

• Restrict the tool access.

• Train the users.

• Monitor the usage regularly.

How transparent is the data sharing process?

Data sharing requires your explicit acknowledgment before you connect NetSuite with a third-party LLM. You are responsible for maintaining control over what data is shared and how the integration is used within your organization.

What are the performance or data limits?

From NetSuite, concurrency limits apply, which are shared with other integrations.

The AI client may generate multiple requests per action and if too many requests are sent, and the limits are exceeded, the requests will fail with "Too Many Requests" error. You can retry, but it is best to review your account's concurrency limit. The administrator can also allocate limits per integration record, and the remaining capacity is shared. For more information, see NetSuite AI Connector Service And Concurrency Governance.

Regarding reports, only date filters and subsidiary filters are supported. Accounting periods are not supported and subsidiaries affect the report results. For more information see Report Tools.

AI clients have their own limitations. They usually support up to ~5,000 rows per call. Also, token limits may apply, and large queries may cause timeout.

How can I monitor and audit usage?

To monitor and audit usage, you can go to the Execution Log of the integration record. The Execution Log entries are filtered by integration record and include details such as HTTP method and URL path, which help with deeper troubleshooting and auditability. It shows:

• Request time

• Duration

• Status

• User

• HTTP response

• Request details

For more information, see NetSuite AI Connector Service (MCP) Execution Log

I added permissions but cannot connect. What should I check?

Check the following:

• Permissions are assigned to the role.

• The correct permission is used (OAuth 2.0 Access Tokens and not Access Tokens).

• The role is not administrator and does not have administrator rights.

• Required features are enabled.

• The SuiteApp is installed.

I don't see any tools. What should I check?

Make sure:

• The NetSuite AI Connector is properly connected.

• You are logged in with the correct role.

• The SuiteApp (for example, MCP Standard Tools) is installed.

• Your role has the required permissions.

• The access to the SuiteApp folder is not restricted.

Why does the AI client (for example Claude) say I'm connected but the connector is disconnected?

Some possible causes for this issue are:

• You are using incorrect endpoint and your URL is missing the /all.

• The SuiteApp is not installed.

• The role you used to log in is not properly configured.

How do I verify everything is working?

For the NetSuite AI Connector Service to work properly, check that:

• All required features are enabled.

• All correct permissions are assigned.

• The AI client is calling the correct URL.

• An integration record has been created.

• The tools appear in the AI client.

• The Execution log shows activity.