System Requirements Document (SRD)

Project Name: amber-news

1. Introduction

The amber-news project is designed to address the growing challenge of misinformation by providing users with a robust fake news analyzer. This system will empower individuals, journalists, and researchers in India to assess the credibility of news articles, links, or text. By leveraging advanced algorithms and APIs, amber-news aims to deliver accurate and actionable insights into the reliability of information.

This document outlines the system requirements for amber-news, incorporating the latest updates requested by Anushka Bisney.

2. System Overview

amber-news will be an IT solution that enables users to input news articles, links, or text for analysis. The system will evaluate the credibility of the content using pre-resolved APIs supervised by advanced algorithms. The results will be presented in an intuitive and detailed format, including trust scores, flagged issues, and source reliability metrics.

Key features include:

• User-friendly input methods for text, links, or files.
• Advanced credibility analysis powered by AI models and APIs.
• A visually appealing results page with actionable insights.

The system will cater to users in India, considering locale-specific factors such as language preferences, timezone (IST), and regional news sources.

3. Functional Requirements

Story Points:

• As a User, I should be able to input links, text, or files for analysis.
• As a User, I should receive a trust score and flagged issues for the analyzed content.
• As a User, I should be able to view the reliability metrics of the source.
• As an Admin, I should be able to manage the APIs used for analysis.
• As an Admin, I should be able to monitor system performance and user activity.
• As a Guest, I should be able to explore a demo of the fake news analyzer without registration.
• As a Developer, I should be able to integrate pre-resolved APIs supervised by the system.

4. User Personas

Admin:

• Responsible for managing APIs, monitoring system performance, and ensuring data integrity.

User:

• Everyday readers, journalists, and researchers who want to analyze the credibility of news articles or text.

Guest:

• Individuals exploring the system without registration, accessing limited demo features.

Developer:

• Technical personnel integrating APIs and maintaining the backend infrastructure.

5. Visuals Colors and Theme

Color Palette:

• Background: #F5F5DC (Beige)
• Surface: #FFFFFF (White)
• Text: #2F4F4F (Dark Slate Gray)
• Accent: #FF4500 (Orange Red)
• Muted Tones: #B0C4DE (Light Steel Blue)

The theme will evoke trust and clarity, with a clean and professional aesthetic suitable for analyzing news credibility.

6. Signature Design Concept

Concept: Interactive Newsroom Dashboard

The homepage will resemble a sleek, interactive newsroom dashboard. Users will be greeted with a dynamic interface featuring:

• Real-time animations: A rotating globe showcasing trending news topics and regions.
• Interactive elements: Users can drag and drop links or text into a central "Analyze Now" box, which lights up with a scanning animation.
• Micro-interactions: Hovering over sections will reveal tooltips with explanations of features.
• Dynamic trust score visualization: Results will appear as a circular gauge that fills with color based on the trust score, accompanied by animated flags for issues detected.

This design will make the system engaging and memorable, ensuring users feel empowered to tackle misinformation.

7. Non-Functional Requirements

• The system must support high concurrency to handle multiple users simultaneously.
• Response time for analysis should not exceed 5 seconds.
• The system must comply with data privacy regulations in India.
• Ensure scalability to accommodate future expansions, such as additional languages or regions.

8. Tech Stack

Frontend:

• React for Web

Backend:

• Python
• FastAPI

Database:

• MySQL (with Alembic for migrations)

AI Models:

• GPT 5.4 for user-friendly responses
• Claude 4.6 Opas for academic or coding work
• Google Nano Banana for image generation

AI Tools:

• Litellm for LLM Routing
• Langchain

Local Orchestration:

• Docker
• docker-compose

Server-Side Orchestration:

• Kubernetes

9. Assumptions and Constraints

Assumptions:

• Users will primarily input content in English or Hindi.
• The system will initially focus on news sources relevant to India.
• APIs for analysis will be pre-resolved and supervised for accuracy.

Constraints:

• The system must operate within the constraints of Indian data privacy laws.
• Limited initial support for regional languages beyond Hindi and English.

10. Glossary

• Fake News Analyzer: A tool designed to assess the credibility of news articles, links, or text.
• Trust Score: A numerical representation of the reliability of analyzed content.
• Flagged Issues: Specific problems or concerns identified during the analysis.
• API Supervisor: A system or individual responsible for overseeing the APIs used for analysis.
• Concurrency: The ability of the system to handle multiple users or processes simultaneously.

End of Document