hackathon-team-5

Web Error Handling and Prioritisation

Flow:

Frontend

On the landing page, our application accepts the website link to inspect. It passes this url to NextJs API routes where Puppeteer can take an action over that.

Backend

Puppeteer runs a separate instance of the browser with that website and it has full control over that instance. Puppeteer is automating the flow of operations and meanwhile in this flow, it can also record various errors occurring in the website. At the end of the flow, Puppeteer sends the error logs to Ollama by categorizing them by the webpage Puppeteer has visited. After receiving the response containing solutions of the errors from Ollama, NextJs calls a ROR API to save the data in DB for further use cases. It returns the webpage wise data containing the errors along with their solutions.

LLM

Responsible for giving the solutions of the errors passed to it.

#Setup

System Requirements

Frontend (Next.js)

• Node.js 18+
• npm or yarn
• Ollama installed locally

Backend (Rails)

• Ruby 3.0.0
• PostgreSQL
• Docker (optional)

Setup Instructions

Backend Setup (Rails)

1. For backed we have a docker setup, run docker-compose up --build once it is successfully you can access rails server on localhost:3005

Frontend setup ( nextjs and ollama )

1. Install ollama following

• https://hub.docker.com/r/ollama/ollama
• we are using llama3.2:1b in your code base

2. For nextjs

• Make sure you have nodejs installed on your system with latest lts tag
• Install dependencies for nextjs using npm install
• Once Installed start nextjs server using npm run dev