-
-
Notifications
You must be signed in to change notification settings - Fork 1
Basic Chat Operations
RenzMc edited this page Sep 6, 2025
·
1 revision
This guide explains how to use the basic chat functionality of the Acode AI CLI Assistant Plugin.
There are several ways to open the AI assistant:
-
Windows/Linux:
Ctrl + Shift + A -
Mac:
Cmd + Shift + A
- Open Acode
- Tap the AI CLI Assistant icon in the sidebar
The chat interface consists of three main components:
- Shows the conversation history between you and the AI
- Displays both your queries and AI responses
- Supports markdown formatting for rich text
- Auto-scrolls to show the latest messages
- Where you type your queries
- Supports multi-line input (use Shift + Enter for new lines)
- Has a placeholder with usage instructions
- Send Button: Submit your query to the AI
- Stop Button: Cancel ongoing AI generation
- New Chat Button: Start a fresh conversation
- History Button: Access previous conversations
You can ask the AI anything related to coding:
How do I create a React component?
Explain what a promise is in JavaScript
What's the difference between let and const?
For more specific coding assistance:
How can I optimize this function?
Explain this SQL query
Convert this Python code to JavaScript
Ask about your entire project:
How should I organize this project?
What's the best way to structure these files?
Suggest improvements to my architecture
AI responses are formatted with:
- Markdown support: Code blocks, lists, headers, etc.
- Syntax highlighting: For code snippets in responses
- Copy buttons: Easy copying of code blocks
- Smooth animations: For better visual experience
After receiving a response, you can:
- Copy code blocks: Click the copy button on any code block
- Copy entire messages: Use the menu option to copy all messages
- Export conversations: Save chats for future reference
The plugin automatically saves your conversations:
- Click the History button in the chat header
- Browse through previous conversations
- Click on any conversation to load it
- Delete conversations: Use the X button next to each history item
- Clear all history: Use the menu option to clear all chat history
- Export conversations: Save chats as JSON files for backup
- Conversations are saved locally on your device
- Each conversation is associated with its first query
- History is encrypted along with API keys for security
To start a fresh conversation:
- Click the New Chat button in the chat header
- Or use the menu option "Clear Chat History"
- The message display area will be cleared
- Previous conversation history is saved automatically
- AI responses appear character by character for a natural feel
- You can stop generation at any time using the Stop button
- Visual indicators show when the AI is thinking
The AI automatically has context about:
- Current file name and extension
- Current file content
- Cursor position
- Project structure
Click the menu icon (three dots) in the chat header to access:
- Clear Chat History: Start with a clean slate
- Export Conversation: Save the current chat
- Copy All Messages: Copy entire conversation to clipboard
- Bulk Operations: Access file management tools
- Project Organization: Get AI-powered project structure suggestions
- Settings: Adjust plugin configuration
Change the AI model without switching providers:
- Click the menu icon
- Select "Model"
- Choose from available models
- The conversation continues with the new model
- Be specific about what you need
- Include relevant context when possible
- Ask one question at a time for clearer responses
When asking about code:
- Include the relevant code snippet
- Mention the programming language if not obvious
- Describe what you're trying to achieve
- You can ask follow-up questions in the same conversation
- The AI remembers previous messages in the conversation
- This helps provide more contextually relevant responses
- Check your internet connection
- Verify your API key is valid
- Ensure the selected AI provider is functioning
- Some providers may be slower than others
- Large context can slow down responses
- Try switching to a faster model or provider
- AI models can sometimes provide incorrect information
- Verify critical code before implementing
- Try rephrasing your query for better results
- Some complex markdown may not render perfectly
- Code blocks should always display correctly
- Report rendering issues on GitHub
To verify chat functionality:
- Open the AI assistant
- Send a simple test query
- Wait for response
- Check that formatting and code blocks display correctly
If you encounter issues with chat operations:
- Check the Common Issues documentation
- Visit our GitHub Issues page
- Join our Discussions community for support
- Home
- Getting Started
- Usage Guides
- Advanced Features
- Developer Docs
- Troubleshooting