> ## Documentation Index > Fetch the complete documentation index at: https://user-docs.seaticket.ai/llms.txt > Use this file to discover all available pages before exploring further. # Chat with AI AI Chat is an assistant built into SeaTicket that helps you complete tasks, answer questions, and find relevant information. Unlike a traditional chatbot, AI Chat works directly with your tickets, issues, and documentation. This allows it to understand context, analyze problems, and provide more relevant and actionable answers. # What You Can Do with AI Chat AI Chat supports your daily work with these core capabilities: ### Answer Questions with Context AI Chat understands the content of your tickets, issues, and documents, and can answer questions like: * What is happening in this issue? * What is the root cause? * What should I check first? ### Troubleshoot Problems AI Chat helps you diagnose technical or workflow issues by: * Analyzing error messages and logs * Suggesting possible causes * Recommending next troubleshooting steps ### Search Documentation and Past Issues AI Chat automatically: * Retrieves relevant documentation * References similar historical issues * Provides supporting context and insights ### Build Knowledge for Your Team Turn useful conversations into reusable knowledge by: * Saving solutions as knowledge records * Sharing insights with your team * Reducing repeated problem-solving # How AI Chat Works AI Chat generates responses by combining: * Current ticket or issue context * Attached documents and knowledge base content * Historical issues and discussions * Connected external or internal sources This approach grounds AI Chat's answers in your actual workflows and data. Image ## Customizing AI Behavior with a Project Prompt You can provide the AI with background information about your project, so it responds with better context across all conversations. **To set a Project Prompt:** 1. Go to **Settings** in your SeaTicket project 2. Locate the **Project prompt** section 3. Click **Edit** 4. Enter background information about your project — for example, your product description, project context, or terminology the AI should be aware of 5. Click **Submit** Once set, this prompt is applied automatically to all AI conversations within the project. You do not need to repeat this context in every chat. Untitled Design (6) **Tips for writing an effective Project Prompt:** * Describe what your product or service does * Include key terms or abbreviations your team uses * Mention the type of issues or tickets your team typically handles The Project Prompt helps AI Chat give more relevant, accurate answers without requiring you to re-explain your context each time. ## Ways to Start AI Chat You can start AI Chat from either a ticket or an external issue. ### Start from an Issue 1. Go to **Connections** 2. Open a connected source (for example, GitHub) 3. Select an issue 4. Right-click the issue 5. Choose **Chat issue** AI Chat will open with the issue content as context. You can ask questions like: * What is the root cause? * Has this been solved before? * Summarize the discussion Image ### Start from a Ticket 1. Go to the **Ticket list** (All tickets or My tickets) 2. Select a ticket 3. Right-click the ticket 4. Choose **Chat ticket** AI Chat will open with the ticket information as context. You can ask questions like: * How is this problem usually solved? * Summarize the issue * What should I check first? Image ## Using the Chat Interface After you start a chat, you'll see the Chat page with these key elements: * **Chat Panel** — Ask questions and read responses. The AI generates answers based on your ticket or issue content. You can ask follow-up questions, and the AI maintains the conversation thread. * **Add Tickets and Docs** — Attach relevant tickets or knowledge base articles as context so the AI can analyze them. * **Model Selection** — Choose an AI model based on task complexity: * `gemini-3-flash` or `gemini-3.1-pro` for general Q\&A and fast responses * `gpt-4o-mini` for complex analysis and multi-step reasoning Image ## Generating and Using Documents During a conversation, AI Chat can generate structured Markdown files that you can: * Preview * Copy * Download * Save for future reference Common use cases include troubleshooting documentation, incident reports, and knowledge base articles. Image ## Managing Chat History All chats are automatically saved in your **Mine** or **Team** histories. For each chat, you can: * **Rename** — Update the chat title for easy reference * **Share within team** — Allow colleagues to view and use the AI discussion * **Delete** — Remove a chat from history if no longer needed Image