DevInsight AI is an advanced, production-grade intelligence platform that analyzes local and remote repositories, parses technology dependencies, runs automated security/SEO audits, and evaluates live hosted website PageSpeed metrics (using Google Lighthouse curves).
- 💻 Deep Codebase Auditing: Analyze repositories locally or clone them remotely via Git. Auto-extract libraries, database schemas, API routes, and code-level concerns.
- 🤖 RAG Chat Coding Assistant: Ask architecture questions, explain complex code, or design database schemas directly with a local-first RAG vector index.
- 📈 Google Lighthouse Emulator: Evaluate hosted URLs directly from the dashboard:
- Performance: Mobile and desktop scoring curves matching Google PageSpeed Insights LCP/FCP benchmarks.
- Accessibility: Audits HTML language attributes, semantic structural elements, image alt text compliance, and form labels.
- Best Practices: Evaluates HTTPS secure sockets, response headers, and target-blank link safety.
- SEO: Validates indexing keywords, title and meta lengths, viewport layouts, and H1 distribution.
- 🔍 W3Techs Tech Stack Scanner: Parses live site response headers and source structures to extract backend languages (PHP, Python, ASP.NET), client libraries (React, Next.js, jQuery), and web servers (Nginx, Apache, Cloudflare).
- 🐳 Dockerized Development: Zero dependency setup with single-command Docker environments.
graph TD
User([User]) -->|Inputs Git URL / Local Path| DevInsight[DevInsight AI Portal]
User -->|Inputs Hosted URL| Auditor[Lighthouse Auditor]
subgraph Repo Analysis Engine
DevInsight -->|Git Service| Cloner[Cloner / Reader]
Cloner -->|AST Parser| Parser[Tree-Sitter / Language Extractor]
Parser -->|RAG Vectorization| ChromaDB[(Chroma Vector DB)]
ChromaDB -->|LLM Prompt Context| AI[AI Advisor / Advisor RAG]
end
subgraph Host Audit Engine
Auditor -->|HTTP Scraper| Scraper[Target Site Scraper]
Scraper -->|W3Techs Scanner| TechDetect[Technology Detector]
Scraper -->|SEO & Alt audits| Landmarks[Landmarks & Heuristics Evaluator]
Landmarks -->|Lighthouse Curves| Gauges[Mobile & Desktop SVG Gauges]
end
AI -->|Response| User
Gauges -->|Dashboard Visuals| User
- Backend: Django 5.1, Django Rest Framework, Celery, PostgreSQL / SQLite.
- AI & Search Vector DB: OpenAI, Anthropic, Google Generative AI, ChromaDB, SentenceTransformers.
- Parsers: Tree-Sitter AST (Abstract Syntax Tree) Parser, GitPython, PyDriller.
- Frontend: Vanilla CSS3, SVG gauges, and asynchronous AJAX chat handlers.
Make sure you have installed:
- Python 3.13+
- Git
- Docker & Docker Compose (Optional, for containers)
-
Clone the Repository:
git clone https://github.com/dev-amit-dots/DevInsightAI.git cd DevInsightAI -
Setup Virtual Environment:
python -m venv venv # On Windows: .\venv\Scripts\activate # On macOS/Linux: source venv/bin/activate
-
Install Dependencies:
pip install -r requirements.txt
-
Run Database Migrations:
python manage.py makemigrations python manage.py migrate
-
Create Superuser (Optional):
python manage.py createsuperuser
-
Start the Development Server:
python manage.py runserver
Access the dashboard at
http://localhost:8000/.
Run the entire application including Celery workers and Redis caching using Docker Compose:
docker-compose up --buildAccess the application at http://localhost:8000/.
Verify your local installation is fully correct by executing the Django/pytest test suite:
# Run pytest ignoring data volumes
.\venv\Scripts\pytest.exe --ignore=data- Username/Email:
test@mailinator.com - Password:
Test@123
Distributed under the MIT License. See LICENSE for details.
Amit Tiwari
- Backend Developer | Core PHP | Laravel | Python | Django | FastAPI
- GitHub: https://github.com/dev-amit-dots
- LinkedIn: https://linkedin.com/in/amit-tiwari-2866-dev