System Requirements Document (SRD)

Project Name: onyx-prompt

1. Introduction

The onyx-prompt project is an advanced, AI-powered application designed to revolutionize both prompt generation and image intelligence. This document outlines the system requirements for the upgraded version of the application, which now features a two-tab structure: Prompt Generator and Image Intelligence.

The Prompt Generator tab provides users with a structured, customizable, and high-quality prompt creation tool, while the Image Intelligence tab leverages cutting-edge AI capabilities to offer unparalleled reverse image search and analysis functionality.

This SRD has been tailored for John Doe, based in the United States, with locale-specific considerations such as US time zones, English language defaults, and accessibility for a broad audience of professionals, researchers, and general users.

2. System Overview

The onyx-prompt application is a two-tab platform that combines advanced AI technologies to deliver the following:

1. Prompt Generator Tab:

   • A 7-step configuration process for creating master prompts tailored to specific use cases (e.g., creative, technical, business, research).
   • Modular toggles for enhancing prompts with features like Chain of Thought, Expert Persona, and Error Fallbacks.
   • AI-generated outputs structured into sections such as Role Declaration, Mission Brief, and Activation Phrase.
   • Quality scoring for generated prompts, with animated visual feedback.

2. Image Intelligence Tab:

   • Powered by Claude Vision AI, offering deep understanding and analysis of images.
   • Features include OCR, origin detection, deep insights, AI-optimized search queries, and integration with six reverse search engines.
   • One-tap functionality for copying images to the clipboard for seamless use across platforms.

The application is designed for deployment as a web app and can be converted into an APK for mobile use via simple steps.

3. Functional Requirements

Prompt Generator Tab

• As a User, I should be able to configure a 7-step process to generate master prompts.
• As a User, I should be able to select the type of prompt (e.g., Creative, Technical, Business, Persona, Research, Marketing, Image Gen, Workflow).
• As a User, I should be able to target specific AI models (e.g., Claude, ChatGPT, Gemini, Midjourney).
• As a User, I should be able to define tone, output format, and goals with extra context.
• As a User, I should be able to toggle 10 prompt enhancers on/off (e.g., Chain of Thought, Expert Persona, Anti-Pattern Warnings).
• As a User, I should receive a fully structured master prompt with sections like Role Declaration, Mission Brief, and Activation Phrase.
• As a User, I should see animated quality scores for Specificity, Completeness, and AI Compatibility.
• As a User, I should be able to copy the generated prompt with one click.

Image Intelligence Tab

• As a User, I should be able to upload an image for analysis.
• As a User, I should receive OCR results that extract all visible text from the image.
• As a User, I should see origin detection results, including guesses about the image's source (e.g., website, brand, creator type).
• As a User, I should receive deep insights about the image, such as unique observations and context.
• As a User, I should be presented with 5 AI-optimized search queries for further exploration.
• As a User, I should be able to perform reverse image searches using six integrated engines (Google Lens, TinEye, Yandex, Bing, SauceNAO, IQDB).
• As a User, I should be able to copy the image to the clipboard with one tap.

4. User Personas

1. General User

• Description: Individuals seeking intuitive tools for generating prompts or analyzing images.
• Goals: Simplify tasks, save time, and enhance productivity.

2. Researchers

• Description: Academics, analysts, or professionals who need detailed insights and accurate data for their work.
• Goals: Access advanced AI tools for image analysis and prompt generation to support research.

3. Content Creators

• Description: Writers, marketers, and designers who require creative prompts and image insights for their projects.
• Goals: Generate high-quality prompts and find image sources quickly.

5. Visuals Colors and Theme

Color Palette

The onyx-prompt app will feature a sleek, modern, and professional design with the following unique color palette:

• Background: #1A1A2E (Deep Midnight Blue)
• Surface: #16213E (Dark Navy)
• Text: #EAEAEA (Soft White)
• Accent: #F05454 (Vibrant Coral Red)
• Muted Tones: #6C757D (Muted Gray)

6. Signature Design Concept

The onyx-prompt homepage will feature an Interactive AI Universe.

Concept Details:

• Visuals: A dynamic, 3D galaxy map where each star represents a feature or section of the app.
• Interaction: Users can hover over stars to reveal tooltips with feature descriptions, and clicking on a star zooms into that feature's "planet," which displays detailed functionality.
• Animations: Smooth transitions between zoom levels, with stars pulsating subtly to indicate interactivity.
• Color Shifts: The galaxy's background will shift between hues of deep blue and purple, creating a mesmerizing effect.
• Micro-Interactions: Clicking on a star triggers a ripple effect across the galaxy, emphasizing the interconnectedness of features.

This design ensures that the first impression is unforgettable, blending functionality with a sense of wonder and exploration.

7. Non-Functional Requirements

• The app must load within 3 seconds on standard broadband connections.
• The app must be responsive and accessible on both desktop and mobile devices.
• The app must comply with ethical guidelines and avoid promoting harmful or inappropriate content.
• The app must support high-resolution images for analysis without significant performance degradation.

8. Tech Stack

Frontend

• React for Web

Backend

• Python
• FastAPI

Database

• MySQL (with Alembic for migrations)

AI Models

• Claude Vision AI for image analysis
• GPT 5.4 for user-friendly responses

AI Tools

• LangChain
• Litellm for LLM Routing

Orchestration

• Docker
• Kubernetes

9. Assumptions and Constraints

• The app assumes users have access to high-quality images for analysis.
• The app assumes users will provide detailed input for prompt generation to maximize output quality.
• Deployment assumes the use of platforms like Netlify and Median for APK generation.

10. Glossary

• OCR: Optical Character Recognition, a technology for extracting text from images.
• Reverse Image Search: A process of finding visually similar images or related information based on an uploaded image.
• Prompt Enhancers: Features that improve the quality and specificity of AI-generated prompts.
• Claude Vision AI: An advanced AI model for image understanding and analysis.

End of Document.