The MN Analise extension for Visual Studio Code provides comprehensive code analysis for security vulnerabilities using multiple AI providers. The extension offers a simple and fast way to detect potential security issues in your code and receive detailed explanations for any vulnerabilities identified.
Features
Multi-AI Provider Support:
OpenAI's ChatGPT (GPT-3.5-turbo)
Google's Gemini AI
Local Ollama models (codellama, llama2, mistral, etc.)
Vulnerability Detection: Analyze selected code snippets for potential vulnerabilities
Detailed Explanations: Receive comprehensive explanations for detected vulnerabilities
Interactive Webview Panel: View analysis results in a dedicated panel within VS Code
Language Support: Choose between English and Portuguese (BR) for analysis results
Customizable Settings: Configure API keys, AI providers, and analysis preferences
Dedicated Sidebar: Easy access through the VS Code activity bar with quick actions
Prerequisites
Before using the extension, ensure you have:
Visual Studio Code installed
At least one of the following:
OpenAI API key (for ChatGPT)
Google Gemini API key (for Gemini AI)
Ollama installed locally (for local model analysis)
Installation
Install the extension from the Visual Studio Code marketplace
User Interface
The extension provides a dedicated sidebar accessible through the VS Code activity bar:
Sidebar Features
MN Analise Icon: Located in the VS Code activity bar (left sidebar)
Search Vulnerabilities: Quick access to code analysis functionality
Settings: Direct link to extension configuration
Clean Interface: Simple and intuitive navigation
Activity Bar Integration
The extension adds its own icon to the VS Code activity bar
Click the icon to open the dedicated sidebar panel
Provides quick access to all extension features without using the command palette
Configuration
The extension offers several configuration options to customize your experience:
1. Language Settings
Language: Choose between English and Portuguese (BR) for analysis results
This setting affects the language of vulnerability explanations and interface messages
2. AI Provider Selection
ChatGPT (OpenAI): Use OpenAI's GPT models for analysis
Gemini (Google): Use Google's Gemini AI for analysis
Ollama (Local): Use local Ollama models for offline analysis
3. API Configuration
OpenAI API Key: Required when using ChatGPT provider
Gemini API Key: Required when using Gemini provider
Ollama Settings: Configure model name and endpoint for local analysis
Usage
Method 1: Using the Sidebar Menu
Look for the "MN Analise" icon in the VS Code activity bar (left sidebar)
Click on the icon to open the extension's dedicated sidebar
In the sidebar, you'll find two options:
"Search vulnerabilities" - Click to analyze your selected code
"Settings" - Quick access to extension configuration
Select the code you want to analyze in your editor
Click "Search vulnerabilities" in the sidebar
View the analysis results in the webview panel
Method 2: Using the Command Palette
Configure your preferred settings in the Extension Settings:
Language: Choose your preferred language (English or Portuguese)
AI Provider: Select your AI provider (ChatGPT, Gemini, or Ollama)
API Keys: Set up the corresponding API key(s) for your chosen provider
Ollama Settings: (Optional) Configure Ollama settings if using local models
Open a code file in Visual Studio Code
Select the code snippet you wish to analyze
Press Ctrl+Shift+P (or Cmd+Shift+P on Mac) to open the command palette
Type "Search vulnerabilities" and select the command
View the analysis results in the webview panel
Video Tutorial
Click Here to Dive Into Our Step-by-Step Video Guide.
Release Notes
Version 1.0.0
Initial release of the MN Analysis extension
Version 2.0.1
Added setting to use one's own OpenAI API Key
Version 3.0.0
Added setting to select the language
UI/UX improvements
Version 4.0.0
Added support for Google's Gemini AI
Added setting to specify the AI provider (OpenAI or Gemini)
Version 5.0.0
Added support for local Ollama models
Enhanced configuration options for AI providers
Improved error handling and logging
Added support for multiple language options
Version 5.0.3
Fixed configuration order using numbered prefixes
Language selection now appears first in settings (01.language)
AI provider selection appears second (02.aiProvider)
API keys and Ollama settings follow in logical order
Ensures consistent display order in VS Code settings
