CodeNova.Devs

CodeNova.Devs is an AI-powered developer suite built with a Streamlit frontend and a FastAPI backend. It combines chat, code-generation tools, debugging helpers, API utilities, generators, snippet storage, PDF export, voice features, and SQLite-backed session history in one project.

Current Architecture

This repo no longer runs as a single Streamlit-only app. It is split into:

• main.py - Streamlit frontend
• api.py - FastAPI backend wrapper
• utils.py - shared app logic, model calls, DB helpers, exports, and tool functions
• codenova.db - SQLite database for sessions, API profiles, and snippets

flowchart LR
    U["User"] --> S["Streamlit UI\nmain.py"]
    S --> A["FastAPI Backend\napi.py"]
    A --> L["LangChain + Groq"]
    A --> D["SQLite\ncodenova.db"]
    A --> X["Shared Logic\nutils.py"]

Loading

Features

• AI chat with Groq-backed models
• Session history with PDF and JSON export
• Voice input and text-to-speech support
• Code tools for explain, translate, refactor, security review, quality review, and debugging
• Dev sandbox for code execution, auto-fix, and time-complexity checks
• API suite for request testing, health monitoring, and saved API profiles
• Generators for unit tests, regex, SQL, Dockerfiles, commit messages, and changelogs
• Snippet library stored in SQLite
• Backend health/reconnect handling in the Streamlit UI

Tech Stack

• Python
• Streamlit
• FastAPI
• LangChain
• Groq
• HTTPX / Requests
• SQLite
• Plotly / NumPy
• gTTS / SpeechRecognition
• FPDF

Project Structure

CodeNova.Devs/
├── api.py
├── main.py
├── utils.py
├── requirements.txt
├── .env.example
├── render.yaml
├── codenova.db
└── README.md

Local Development

1. Clone the repo

git clone https://github.com/rudra00434/CodeNova.Devs.git
cd CodeNova.Devs

2. Create and activate a virtual environment

python -m venv .venv

Windows:

.venv\Scripts\activate

macOS / Linux:

source .venv/bin/activate

3. Install dependencies

pip install -r requirements.txt

4. Configure environment variables

Copy .env.example to .env and update the values you need.

Minimum required:

GROQ_API_KEY=your_groq_key_here

5. Start the backend

uvicorn api:app --reload

6. Start the frontend

In a second terminal:

streamlit run main.py

7. Open the app

• Frontend: http://127.0.0.1:8501
• Backend API: http://127.0.0.1:8000
• Swagger docs: http://127.0.0.1:8000/docs

Environment Variables

Variable	Purpose
GROQ_API_KEY	Required on the backend for AI-powered features
CODENOVA_API_BASE_URL	Explicit backend URL for Streamlit, usually http://127.0.0.1:8000 locally
CODENOVA_API_HOSTPORT	Internal Render host/port value used when frontend talks to a private backend
CODENOVA_API_TIMEOUT	HTTP timeout for frontend-to-backend calls
CODENOVA_API_RETRIES	Retry count for API requests
CODENOVA_API_RETRY_BACKOFF	Retry backoff between attempts
CODENOVA_HEALTH_POLL_MS	Health poll interval when backend recovery is enabled
CODENOVA_HEALTH_AUTO_POLL	Enables automatic backend health polling
CODENOVA_DB_PATH	Custom SQLite location for the backend
PORT	Service port used by deployment platforms such as Render

FastAPI API Overview

Major route groups exposed by api.py:

• Core: /, /health, /meta
• Sessions: /sessions, /sessions/{session_id}/history, /sessions/{session_id}/export.pdf
• Chat: /chat, /tts
• Code tools: /tools/explain, /tools/translate, /tools/refactor, /tools/security, /tools/quality, /tools/debug
• Sandbox: /sandbox/run, /sandbox/fix, /sandbox/complexity
• API suite: /http/request, /monitor, /api-profiles
• Generators: /generate/unit-tests, /generate/regex, /generate/sql, /generate/dockerfile, /generate/commit-message, /generate/changelog
• Snippets: /snippets

Deployment on Render

There are two valid deployment patterns now.

Option A: render.yaml (private backend + persistent disk)

The included render.yaml is designed for a more complete setup:

• private FastAPI backend
• public Streamlit frontend
• persistent disk for SQLite

This is the better production layout, but it requires Render features that are not available on the free plan.

Option B: Free Render setup (two public web services)

If you want to stay on the free plan, deploy the services manually:

Backend service

• Type: Web Service
• Build command:

pip install -r requirements.txt

• Start command:

uvicorn api:app --host 0.0.0.0 --port $PORT

• Recommended env vars:

GROQ_API_KEY=your_groq_key_here
PYTHON_VERSION=3.11.11

Frontend service

• Type: Web Service
• Build command:

pip install -r requirements.txt

• Start command:

streamlit run main.py --server.port $PORT --server.address 0.0.0.0

• Recommended env vars:

CODENOVA_API_BASE_URL=https://your-backend-service.onrender.com
PYTHON_VERSION=3.11.11
CODENOVA_API_TIMEOUT=180
CODENOVA_API_RETRIES=0
CODENOVA_HEALTH_AUTO_POLL=false
CODENOVA_HEALTH_POLL_MS=0

Important Render Notes

• Share the frontend URL publicly, not the backend URL.
• On the free plan, services can sleep after inactivity and may need time to wake up.
• SQLite is not persistent on free Render unless you use a persistent disk, so saved sessions/snippets may reset on redeploy or restart.
• For Render health checks, point the backend service to /health.

Dependencies

Current Python dependencies:

• streamlit
• streamlit-lottie
• langchain
• langchain-core
• langchain-groq
• fastapi
• uvicorn
• httpx
• requests
• python-dotenv
• speechrecognition
• fpdf
• gtts
• plotly
• numpy

Notes

• main.py expects a running FastAPI backend.
• If the backend is offline, the Streamlit app falls back to limited metadata mode and API-powered actions will fail until FastAPI is reachable.
• The backend stores app data in SQLite through utils.py.

Maintainer

Built by Rudranil Goswami.