⚡ Local Test Case Generator (BLAST)

A privacy-focused, offline-capable AI tool for generating comprehensive software test cases. Built with the BLAST protocol (Blueprint, Link, Architect, Stylize, Trigger), this tool leverages a local Ollama model to ensure your sensitive requirements never leave your machine.

---

🏗️ Architecture

The application follows a Decentralized Client-Server architecture where the frontend interacts directly with the local AI service, bypassing the need for a heavy backend server.

graph TD
    subgraph "User Environment (Local Machine)"
        Browser[("🌐 Web Browser (Frontend)")]
        ServeScript[("🐍 Python Server (Port 3002)")]
        Ollama[("🦙 Ollama AI Service (Port 11434)")]
    end

    User((👤 User)) -->|"1. Opens http://localhost:3002"| Browser
    ServeScript -->|"2. Serves HTML/CSS/JS"| Browser
    
    Browser -->|"3. Users Enters Requirements"| Browser
    Browser -->|"4. Sends Prompt (POST)"| Ollama
    Ollama -->|"5. Returns Test Cases (JSON)"| Browser
    
    Browser -->|"6. Renders Cards & PDF"| User

Loading

🖼️ Visual Flow (ASCII)

+-------------------+           +----------------------------+
|   👤 User         |           |   💻 Local Machine         |
+-------------------+           |                            |
          |                     |    +------------------+    |
   1. Open Browser              |    | 🌐 Browser (UI)  |    |
          v                     |    | (index.html)     |    |
+-------------------+           |    +--------+---------+    |
| 🌐 http://localhost| <------->+|            |             |
|       :3002       |     2. Serves Assets    | 3. Send Prompt (POST)
+-------------------+            |    +--------v---------+    |
                                |    | 🦙 Ollama API    |    |
                                |    | (Port 11434)     |    |
                                |    +--------+---------+    |
                                |            |               |
                                |     4. JSON Response       |
                                |            |               |
                                |    +--------v---------+    |
                                |    | 📄 JS PDF Gen    |    |
                                |    +------------------+    |
                                +----------------------------+

🧩 Core Components

1. Frontend (UI): Vanilla HTML5, CSS3 (Cyberpunk/Neon Theme), and JavaScript. Handles all logic, state management, and PDF generation (via jsPDF).
2. Server (Host): A simple Python script (serve.py) that uses http.server to serve static files locally.
3. Intelligence (AI): Ollama running the tinyllama model (or any compatible model) to process natural language requirements into structured test cases.

---

📸 Screenshots

Home Page	Generating

Result View	Generated Output

📄 Download Test Case Generated Report

---

✨ Key Features

• 100% Local & Private: No data is sent to the cloud. Perfect for enterprise or sensitive projects.
• Smart PDF Export: Instantly generate timestamped PDF reports (Test_Cases_Report_MM-DD-YYYY...pdf).
• Cyberpunk Speed UI: A high-performance, neon-styled interface designed for focus and speed.
• Multi-Vector Generation: Automatically generates 5-7 detailed test cases (Functional, Edge, Negative) from a single prompt.
• Cross-Platform: Runs on any Windows machine with Python (no heavy backend required).

---

🚀 Getting Started

Prerequisites

1. Ollama: Download and install from ollama.com.
2. TinyLlama Model: Run ollama run tinyllama in your terminal to pull the model.
3. CORS Configuration:
   • For the browser to talk to Ollama, set the environment variable:
   • PowerShell: [System.Environment]::SetEnvironmentVariable("OLLAMA_ORIGINS", "*", "User")
   • Restart your terminal/computer after setting this.

Installation & Run

1. Clone/Download this repository.
2. Start the Server:
   • Ensure you have Python installed.
   • Open terminal in project directory.
   • Run:

     python tools/serve.py

3. Access the App:
   • Open your browser and navigate to the address shown in terminal (e.g., http://localhost:3002).

---

📂 Project Structure

Project Root/
├── tools/
│   └── serve.py           # Python Web Server
├── architecture/
│   ├── SOP_Generation_Logic.md
│   └── SOP_UI_State.md
├── index.html             # Main Application Structure
├── style.css              # Cyberpunk Visual Styles
├── main.js                # Frontend Logic & API Integration
├── BLAST.md               # Development Protocol
├── README.md              # Documentation
└── Test_Case_Generated_Report.pdf # Sample Output

---

🔧 Troubleshooting

• "Ollama connection failed": Ensure Ollama is running (ollama serve) and the OLLAMA_ORIGINS variable is set correctly.
• 404 Errors: Ensure you are running the serve.py script from the Project Root directory.

---

📜 License

This project is licensed under the MIT License - see the LICENSE file for details.

---