Skip to content

FEATURES.md

Latest commit

 

History

History
171 lines (131 loc) · 5.09 KB

FEATURES.md

File metadata and controls

171 lines (131 loc) · 5.09 KB

DevX Extension - New Features Guide

🎯 Overview

This extension provides an intelligent learning environment with adaptive AI assistance, inline feedback, and conversational guidance.

✨ Key Features

1. 💬 Chat with Sensei

Open a conversational chat panel to ask Sensei anything about coding.

How to use:

  • Press Ctrl+Shift+P (or Cmd+Shift+P on Mac)
  • Search for "Chat with Sensei"
  • Or run command: DevX: 💬 Chat with Sensei

Features:

  • Real-time conversational AI assistance
  • Context-aware responses (knows your current code)
  • Chat history for reference
  • Beautiful webview interface
  • Optional - can be closed when not needed

2. 🧙‍♂️ Inline Sensei Feedback (V1.1 Style)

Sensei provides supportive, inline feedback automatically when you pause coding.

How it works:

  • Pause your cursor for 2 seconds on any line
  • Sensei analyzes your code and provides supportive feedback
  • Messages appear inline after your code like:
    • "looks pretty clean, amazing coder! 💯"
    • "nice approach here! ✨"
    • "great logic flow! 🎯"
    • "consider edge cases here 🤔"

Features:

  • Non-repetitive: Shows only once per line
  • Auto-clears: Disappears after 6 seconds
  • Contextual: Analyzes surrounding code
  • Level-adaptive: Feedback matches your skill level

3. 📊 Improved API Error Handling

Better error dialogs when AI providers have issues.

Error Types:

  1. Network Timeout - Cannot reach provider
  2. Empty Response - Model unavailable/overloaded
  3. API Limit Exhausted - Too many requests

Features:

  • Modal dialogs with clear error explanations
  • Quick actions: Switch to Gemini, Change Provider, Retry
  • Provider recommendations when APIs are exhausted
  • Auto-switching option to Gemini (most reliable)

Example:

⚡ OpenRouter API Limit Exhausted

You've made too many requests or exhausted your quota.

[Switch to Gemini] [Try Different Model] [View Alternatives]

4. 🎓 Adaptive Learning System (3 Levels)

Beginner Level:

  • Full working code with line-by-line comments
  • Detailed explanations
  • Step-by-step guidance

Intermediate Level:

  • Strategic hints with TODO/HINT comments only
  • No full code solutions
  • Encourages independent problem-solving

Pro Level:

  • Architecture and design pattern suggestions
  • High-level guidance
  • Assumes strong fundamentals

5. 🔒 Learning Mode Restrictions

Permanently Blocked:

  • Copy (Ctrl+C)
  • Paste (Ctrl+V)
  • Cut (Ctrl+X)

Toggle-able:

  • Tab key - Press Ctrl+Alt+G to toggle (hidden feature)
  • Shows notification: "✅ Tab key ENABLED" or "🚫 Tab key DISABLED"

6. 🎯 Contextual Guidance System

Request Guidance:

  • Command: DevX: Request Learning Guidance
  • Provides level-appropriate help based on:
    • Current code context
    • Line you're stuck on
    • Your learning level
    • Recent activity

Idle Detection:

  • Triggers after inactivity (15s/30s/120s based on level)
  • Suggests next steps conversationally

Stuck Detection:

  • Detects when you're stuck on a line for 30+ seconds
  • Offers: "Yes, give me a hint" / "Let me think more" / "Explain this line"

🚀 Commands Reference

Command Keybinding Description
Chat with Sensei - Opens chat panel
Ask Sensei Ctrl+I / Cmd+I Get contextual help
Request Guidance - Request learning guidance
Change Level - Switch between beginner/intermediate/pro
View Stats - See learning progress
Toggle Tab Ctrl+Alt+G Enable/disable tab key
Select Model Ctrl+J / Cmd+J Choose AI model
Change Provider - Switch AI provider

🔧 Configuration

AI Providers:

  • Gemini (Recommended) - Free, 1500 requests/day
  • Local Ollama - Unlimited, offline, privacy-focused
  • OpenRouter - Pay-as-you-go, many models
  • Ollama Cloud - Cloud-hosted Ollama models

Learning Levels:

  • Auto-adapts based on performance
  • Manual override available via "Change Level" command

💡 Tips

  1. Use Chat for Queries: Open the chat panel when you have questions
  2. Let Sensei Guide: Pause coding for 2s to get inline feedback
  3. Request Help When Stuck: Use "Request Guidance" command
  4. Switch Providers: If one API is down, switch to Gemini
  5. Toggle Tab Temporarily: Use Ctrl+Alt+G if you need tab completion

🐛 Troubleshooting

Inline feedback not showing:

  • Make sure your cursor is on a line with code (not empty)
  • Pause for a full 2 seconds
  • Check if it already showed on that line (non-repetitive)

API errors:

  • Use modal dialog options to switch providers
  • Gemini is most reliable (free tier)
  • Local Ollama requires installation

Chat not responding:

  • Check API configuration
  • Ensure AI provider is selected
  • View console for errors

📝 Notes

  • All features are designed to encourage learning through typing
  • Copy/paste restrictions help build muscle memory
  • Inline feedback is supportive and encouraging
  • Chat panel is optional - use when needed
  • API error handling guides you to working alternatives

Version: 1.1.0
Last Updated: 2024