⬡ QyverixAI
Debug. Understand. Ship faster.
An open-source AI-powered developer assistant that helps beginners understand code, detect bugs, and improve code quality - with plain-English explanations and actionable suggestions.
Live Demo · API Docs · Contributing Guide · Good First Issues
Many beginners struggle with reading error messages, understanding what their code does, or knowing how to improve it. QyverixAI solves this with a clean workspace where you paste code and instantly get:
- A plain-English explanation - what the code does and why
- A structured bug report - what's wrong, which line, and how to fix it
- Improvement suggestions - style tips, best practices, and a quality score
No account required. No API key needed to get started. Fully open source.
| Feature | Description |
|---|---|
| Code Explanation | Language detection, summary, key observations, complexity estimate |
| Bug Detection | 15+ rule patterns: ZeroDivisionError, bare excepts, hardcoded secrets, eval(), and more |
| Improvement Suggestions | Pythonic patterns, documentation gaps, dead code, quality score 0–100 |
| Full Analysis Endpoint | One call — all three analyses combined |
| File Upload | Drag-drop or upload .py, .js, .java, .ts, .cpp files |
| Dark / Light Mode | Persisted theme preference |
| Query History | Last 50 queries saved locally |
| Saved Favorites | Bookmark important results |
| Download Results | Export analysis as .txt |
| LLM-Ready | Drop-in OpenAI-compatible API support via env vars |
| Swagger Docs | Full interactive API documentation at /docs |
git clone https://github.com/imDarshanGK/AI-dev-assistant.git
cd AI-dev-assistantcd backend
pip install -r requirements.txtcd backend
uvicorn app.main:app --host 0.0.0.0 --port 8000 --reloadhttp://localhost:8000/app/
cd backend
pytest -qAll endpoints accept POST with JSON body { "code": "..." }.
Optional: { "code": "...", "language": "python" } to override language detection.
Returns a plain-English breakdown of the code.
{
"language": "Python",
"summary": "This beginner-level Python snippet defines a reusable function...",
"key_points": [
"The code is written in Python with 2 lines.",
"Defines 1 function: add.",
"No loops or conditionals found."
],
"complexity": "Beginner"
}Returns detected issues with line numbers and fix suggestions.
{
"issues": [
{
"type": "Hardcoded Secret",
"line": 3,
"description": "Hardcoded password found in code.",
"suggestion": "Use environment variables: os.getenv('PASSWORD')",
"severity": "error"
}
],
"summary": "Found 1 issue. 1 error(s), 0 warning(s).",
"clean": false
}Returns improvement suggestion cards.
{
"suggestions": [
{
"category": "Documentation",
"description": "Add docstrings to your functions.",
"example": "def greet(name: str) -> str:\n \"\"\"Return a greeting string.\"\"\"",
"priority": "medium"
}
],
"overall_score": 80,
"next_step": "Great code! Consider adding tests next."
}All three analyses in one response.
{
"provider": "rule-based",
"model": "built-in",
"explanation": { ... },
"debugging": { ... },
"suggestions": { ... }
}AI-dev-assistant/
├── backend/
│ ├── app/
│ │ ├── main.py # FastAPI app, middleware, routes
│ │ ├── schemas.py # Pydantic request/response models
│ │ ├── routers/
│ │ │ ├── analyze.py # POST /analyze/
│ │ │ ├── debugging.py # POST /debugging/
│ │ │ ├── explanation.py # POST /explanation/
│ │ │ └── suggestions.py # POST /suggestions/
│ │ └── services/
│ │ ├── code_assistant.py # Rule-based analysis engine
│ │ └── ai_provider.py # LLM abstraction layer
│ ├── requirements.txt
│ └── tests/
│ └── test_endpoints.py # Full test suite (pytest)
├── frontend/
│ ├── index.html # Main UI — no build step
│ ├── style.css # Dark/light theme, responsive
│ └── script.js # All interactivity
├── .env.example # Environment variable reference
├── Dockerfile # One-service Docker build
├── render.yaml # Render deploy blueprint
└── README.md
We welcome contributions from all levels. See CONTRIBUTING.md for guidelines.
New to open source? Check good first issue.
Have questions? Ask in GitHub Discussions.
---
## Deployment
### Render (Recommended - Free Tier)
1. Fork this repository
2. Connect it to [Render](https://render.com)
3. Use the `render.yaml` blueprint - one service, zero config
4. Your app will be live at `https://your-service.onrender.com`
The same service serves the frontend at `/app/` and the API from `/`.
### Docker
```bash
docker build -t qyverixai .
docker run -p 8000:8000 qyverixai
QyverixAI works out of the box with its rule-based engine. To enable richer AI-powered analysis, set these on your backend host:
LLM_ENABLED=true
LLM_API_KEY=sk-your-key-here
LLM_BASE_URL=https://api.openai.com/v1
LLM_MODEL=gpt-4o-miniCompatible with OpenAI, Groq, Together AI, Ollama, and any OpenAI-compatible endpoint.
⚠️ Never commit API keys. Use environment variables or your host's secrets manager.
QyverixAI is a GSSoC 2026 project - beginner contributors are warmly welcome!
# Fork → Clone → Branch → Code → Test → PR
git checkout -b feat/your-feature-name
pytest -q # All tests must passSee CONTRIBUTING.md for the full workflow.
Look for issues labeled good first issue:
- Add support for a new language pattern in the debug engine
- Improve explanation key points for a specific language
- Add a new suggestion rule
- Write tests for edge cases
- Improve frontend accessibility
| Layer | Technology |
|---|---|
| Backend | FastAPI, Pydantic v2, Python 3.12 |
| Frontend | HTML5, CSS3, Vanilla JS (no build step) |
| Testing | Pytest, FastAPI TestClient |
| Deploy | Docker, Render |
| CI | GitHub Actions |
- Rule-based code explanation engine
- Bug detection with 15+ patterns
- Improvement suggestions with quality score
- Full-analysis combined endpoint
- Dark/light theme, file upload, history, favorites
- LLM provider abstraction layer
- Language-specific analyzers (Python, JS, Java, Go)
- Per-user history with database backend
- VS Code extension
- AI-powered explanations (LLM integration GA)
- Multi-file analysis support
MIT © Darshan G K
Built with ♥ for the open source community · GSSoC 2026