# AI-assisted integration (Beta) Use AI assistance to speed up the integration process. {% hint style="info" %} **Requirements**: 1. Project built with Unity. 2. The project does not have devtodev integration yet. \ Projects with poorily integrated events can also participate. 3. AI coding agent that have access to the complete application code. {% endhint %} If you have any questions regarding the integration, let us know by using the [`Contact Us`](https://www.devtodev.com/contact-us) form or reach out to our Customer Success team directly within the platform. We will also greatly appreciate your feedback. \ Please add **AI INTEGRATION** when submitting your request. {% hint style="success" %} We created this manual for [**Cursor**](https://www.cursor.com/) AI code editor but you can try using other AI coding agents that can access all of application code files. {% endhint %} {% embed url="" %} This example is for game application made with Unity, however, the process is similar for other applications types {% endembed %} ## Preparation and prompt ### Get **Unity Integration Guide File** Download the **Unity Integration Guide File** and add it to your Unity project folder. \ We recommend creating a folder called `docs` and placing the guide inside it. This ensures the **AI coding agent** can easily access and use it during analysis. {% file src="" %} Unity Integration Guide File {% endfile %} Open your AI-agent, and select your Unity project folder. The agent will begin indexing all project files, including the markdown guide. Wait a few moments until indexing is complete. ### Create prompt Create a **clear and effective prompt** for the AI-agent.\ The prompt should: * Assign the AI a role (like "You are a product engineer"). * Specify your goal: integrate DevToDev events properly. * Reference the guide using `@devtodev_unity_integration_guide.md`. Copy and paste the prompt text into the AI-agent chat window. You can use our [**prompt**](#get-unity-integration-guide-file) as an example. {% hint style="info" %} The prompt accuracy is estimated at **around 90%**, which means that some devtodev method calls may contain errors. Therefore, it’s recommended that developers review the integration code. {% endhint %} ### Add references to integration guide Attach the guide file inside the prompt. You can do this in two ways: 1. Find all the guide mentions in the prompt and delete that placeholder text. Then, drag the file from the `docs` folder into the chat window. The AI-agent will automatically recognize and link the file. 2. Alternatively, type `@` and start typing "devtodev" β€” the agent will suggest the guide file. Usually it is at the top of the list.
Now click **`Send`** and let the AI coding agent generate the integration. It will: * Analyze your codebase * Understand game logic * Suggest where and how to insert DevToDev analytics events. ## Events generation Check and review the list of events generated by the AI-agent. You can edit the list by removing the unnecessary or duplicate events. You can also ask the AI to revise its suggestions. After you have finished reviewing and modifying the suggested events file, you can proceed to the next step of integration. \ The AI-agent may then generate a new class β€” usually something like `AnalyticsManager` β€” to send events to Devtodev. ## Further integration The coding agent might show some errors if Unity has not recompiled yet. Switch to the **Unity Editor** and wait for it to finish compiling. Once done, the errors in AI-agent will disappear. {% hint style="warning" %} The AI **must not change** your original game logic. It may **add** analytics calls (like `DTDAnalytics.CustomEvent(...)`), but should never modify player behavior, game flow, or state handling. If something looks wrong, ask the AI-agent to revise it. {% endhint %} After generating the integration, the AI-agent will also produce a **documentation file** (like `ANALYTICS_SETUP.md`). It describes: * Where events were added * How to use the generated `AnalyticsManager` class * What next steps are required to finish setup. Follow that instruction carefully to complete your devtodev analytics integration properly. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.devtodev.com/integration/integration-of-sdk-v2/sdk-integration/unity/ai-assisted-integration-beta.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.