Knowivate Autopilot is a VS Code extension that provides local LLM chat capabilities by integrating with Ollama. This extension allows you to chat with AI assistants directly within VS Code while maintaining privacy and performance through local model execution.
Features
Chat with Local AI Models
Privacy-Focused: All interactions stay on your machine with no data sent to external services
Model Selection: Choose from any model available in your local Ollama installation
Streaming Responses: Get real-time responses as the AI generates them
Context-Aware Code Assistance
File Inclusion: Add entire file contents to chat context
Code Selection: Share selected code snippets with the AI
Context Indicators: Visual display of included context count
Modern UI
Multiple Chat Sessions: Maintain separate conversations for different tasks
Markdown Rendering: Rich formatting for AI responses including code syntax highlighting
Command Integration: Easy-to-use buttons for common actions
Requirements
Ollama: You must have Ollama installed on your system
Install using VS Code: Extensions > ... > Install from VSIX
Make sure Ollama is running in the background:
ollama serve
Usage
Access the Extension:
Click on the Knowivate Autopilot icon in the activity bar
The chat interface will appear in the sidebar
Select a Model:
Choose an available model from the dropdown menu
Models are loaded from your local Ollama installation
Chat with the AI:
Type your message in the input box and press Enter or click Send
View AI responses with markdown formatting
Add Code Context:
Click "Include File" to add the current file to the chat context
Select code and click "Include Selection" to add only that portion
Context count indicator shows what's included in your prompt
Manage Sessions:
Create a new chat with the "New Chat" button
Toggle history to switch between different conversations
Commands
This extension contributes the following commands:
knowivate-autopilot.includeFile: Include the current file in chat
knowivate-autopilot.includeSelection: Include selected code in chat
Extension Settings
Currently, all settings are managed through the UI. Future versions will include customizable settings.
Known Issues
Requires Ollama to be running in the background
Limited to models available through Ollama
Fixed system prompt (not yet user-customizable)
Troubleshooting
No Models Available:
Make sure Ollama is running (ollama serve)
Check that you have models installed (ollama list)
Connection Issues:
Verify Ollama is accessible at http://localhost:11434
Check Ollama logs for errors
Performance Issues:
Use smaller models for faster responses
Be mindful of context size when including large files
Privacy
This extension processes all data locally on your machine using Ollama. No prompts, code, or responses are sent to external servers. Your data stays private.
Release Notes
0.0.1
Initial release of Knowivate Autopilot
Integration with Ollama API
Chat interface with markdown rendering
Context awareness for files and code selections
Multiple chat sessions
Model selection
Future Features
Context management panel for viewing/removing context items
Prompt templates for common coding tasks
Message editing functionality
Git integration for including diffs and commits
Custom system prompts
Chat history export/import
More advanced context awareness features
License
This extension is proprietary software developed by Knowivate. All rights reserved.
