🧪 RequirementsLab

AI-Powered Requirement Analysis & Code Generation

RequirementsLab now features two powerful modes for turning ideas into reality:

1. Agent Collaboration Hub (New!): Interactive collaboration with a team of AI Agents.
2. Classic Mode: Structured form-based requirement assessment.

---

---

🚀 Mode 1: Agent Collaboration Hub (Interactive)

Path: http://localhost:5272/agents

A multi-agent system where PM, SA, and PG agents collaborate to turn your ideas into running code.

Workflow Visualized

1. Chat with PM Tell the PM what you want to build. The agents will discuss and verify requirements.

2. Auto-Secretary (Form Mode) Click "📝 Form" to review the structured requirement analysis auto-generated from the chat.

3. Architecture Design The SA Agent analyzes system needs and produces professional Mermaid diagrams.

4. One-Click Execution The PG Agent writes the actual code. Click "📂 Open Folder" to access the generated project.

---

🏛️ Mode 2: Classic Features (Structured)

Path: http://localhost:5272/ (Default)

The original robust workflow for deep feasibility analysis.

• 📝 Requirement Input Form: Structured data collection for project goals and scope.
• 🤖 Dual LLM Architecture:
  • LLM 1: Generates potential solutions.
  • LLM 2: Evaluates feasibility (LLM-as-a-Judge).
• 📊 Feasibility Score: 0-100 automated scoring.
• 📋 Implementation Plan: Generates detailed "Analysis Report" and "Implementation Plan" documents with Gantt charts.

---

🛠️ Setup & Tech Stack

1. Configuration: Copy appsettings.template.json to appsettings.json.

   {
     "AzureOpenAI": {
       "DeploymentName": "gpt-5.2-chat"
     }
   }

2. Run:

   dotnet run

Tech Stack:

• Framework: .NET 10 Blazor Server
• AI Orchestration: Microsoft.Extensions.AI (MAI)
• UI: Markdig (Markdown) + Mermaid.js
• Security: OWASP Top 10 Audited

🐳 Deployment (Docker)

You can containerize and deploy the Agent Collaboration Hub using the included Docker template.

Build & Run

# Build (Local)
docker-compose up --build

Access at http://localhost:5272/agents.

Production Build

docker build -t requirements-lab:latest .
docker run -d -p 80:8080 requirements-lab:latest

📄 License

MIT