# Workflows AI Assistant The Workflows AI Assistant lets you build, update, and debug Workflows from a chat interface that sits alongside the workflow canvas. When you open a Workflow, the AI chat panel appears on the left side of the editor. You can type questions or instructions directly into the chat to build, modify, or troubleshoot your Workflow. The workflow canvas updates in real time as the assistant makes changes. You can collapse the chat panel to focus on the canvas, or expand it when you need help. The canvas stays visually anchored when you toggle the panel. ### Editing Workflows You can ask the AI Assistant to make changes to your Workflow. For example, you could ask to: * Add object tracking to a Workflow. * Make your current Workflow run only if a classification model returns a certain result. * Change the confidence threshold of a model used in a Workflow. The AI Assistant will make changes directly on the canvas. Changes are auto-saved as drafts but are not published to your production Workflow until you explicitly publish. {% hint style="info" %} Blocks created by the AI Assistant may not be fully configured. Click into any block to review and adjust its settings for your project needs. {% endhint %} ### Conversation History The sidebar shows your full conversation history across all Workflows. Selecting a conversation loads the associated Workflow and chat context. You can start a new conversation for the same Workflow using the "New conversation" option. ### Legacy Editor A standalone editor without the AI chat is available as a fallback. You can access it by clicking "Use legacy editor" in the Workflow settings. --- # 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.roboflow.com/workflows/workflows-ai-assistant.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.