Get started with Dynatrace Intelligence agentic and generative AI

• Latest Dynatrace
• How-to guide
• 3-min read
• Updated on Mar 03, 2026

Dynatrace Intelligence agentic and generative AI is enabled on the account level by default, meaning that all your environments automatically have access to it. However, AI functionality must still be enabled on the environment level via the settings page, which offers you full control over how Dynatrace Intelligence agentic and generative AI is enabled and configured in your environment.

Enable Dynatrace Intelligence generative AI on your environment

To enable Dynatrace Intelligence generative AI on your environment

If you can't see the settings page, make sure you have Setting Reader and Setting Writer policies assigned. For more information, see read and write permission.

User permissions

After enabling Dynatrace Intelligence generative AI on the environment level, you'll still need to give access to the various generative AI skills to your users. To do so, you have to bind the group they belong to a policy with the following statement that allows access to generative AI:

For more information on managing your policies, see Manage IAM policies.

Enable agentic AI for Dynatrace Assist

Dynatrace Assist allows you to use Dynatrace agentic AI and MCP tools and capabilities to access and analyze your environment data and use it to perform tasks (such as listing problems or generating and executing DQL queries) in addition to answering general questions about Dynatrace.

To use the agentic Dynatrace Assist, you need to

Dynatrace Assist agentic permissions

You will also need additional permissions for calling the agentic AI tools. For the list of tools and permissions they require, see MCP tools.

PII masking

Agentic Dynatrace Assist doesn't provide any PII masking. In order to protect your data, when Dynatrace Assist detects PII in the user prompt, the request is automatically blocked and the prompt isn't sent to the LLM for processing.

Calling multiple tools

While interacting with Dynatrace Assist in agentic mode, Assist can call up to 10 internal MCP tools per response. If your request requires Dynatrace Assist to call more than 10 tools simultaneously, it'll be unable to complete the interaction.

User-based data access

Dynatrace Intelligence agentic and generative AI respects user privileges and permissions. This means that

Enable document suggestion

Document suggestion is a Dynatrace Intelligence agentic and generative AI skill that allows it to recommend to you troubleshooting guides created in Notebooks and Dashboards based on vector similarity. You can leverage Dynatrace Intelligence agentic and generative AI's document suggestion in Problems to quickly retrieve troubleshooting guides written by you or your team for similar problems and reduce mean time to repair (MTTR).

If you want Dynatrace Intelligence agentic and generative AI to suggest troubleshooting guides for similar or repeatedly occurring problem, you'll need to allow it to search through and index documents created in Notebooks and Dashboards and shared with all users in your environment. To ensure you have full control over the security of your data, this functionality is opt-in and is turned off by default.

To enable document suggestion

By default, Dynatrace Intelligence agentic and generative AI indexes troubleshooting guides every 6 hours.

Semantic vector indexing

Dynatrace Intelligence agentic and generative AI uses semantic vector indexing to suggest relevant troubleshooting dashboards and notebooks shared within the environment. It continuously indexes the content of dashboards and notebooks recognized as troubleshooting guides. When a user accesses the troubleshooting view for a specific problem, generative AI compares the problem description to the indexed data using semantic similarity to suggest the most relevant guides.

This process relies on vector representations of both the problem description and the indexed dashboard or notebook content. The smaller the semantic distance between the problem description and a document, the higher its relevance score. This means it's more likely for a document to be suggested by Dynatrace Intelligence agentic and generative AI as a relevant troubleshooting guide.

Enable environment-aware queries

Environment-aware queries can enrich Dynatrace Intelligence agentic and generative AI with your environment's data. This lets you generate more accurate queries that identify and reference relevant entities, events, spans, logs, and metrics from your environment.

If you want Dynatrace Intelligence agentic and generative AI to be aware of the details and structures of your environment's data, you'll need to allow access to Grail. To ensure you have full control over the security of your data, this functionality is opt-in, and admin users can specify which data tables and buckets Dynatrace Intelligence agentic and generative AI is not allowed to access.

To enable environment-aware queries

It can take up to 24 hours for Dynatrace Intelligence agentic and generative AI to build or amend the semantic index after changes are made. If environment-aware queries are disabled and the semantic index already exists, Dynatrace Intelligence agentic and generative AI purges all environment-specific data within 24 hours, and returns to using publicly available sources for building DQL queries. The semantic index is stored only on your Dynatrace tenant.

To learn more about semantic indexing and environment-aware queries, see Environment-aware queries.

Configure your data access

After enabling environment-aware queries, you'll be able to see the settings for configuring data that Dynatrace Intelligence agentic and generative AI isn't allowed to access.

To configure your data access