Aegis MVP Launchpad is an enterprise-grade client portal and AI agent swarm scope estimator built to showcase top-tier engineering quality, visual excellence, and DevOps practices for an elite MVP Development Service.
This repository demonstrates how a modern MVP development agency operates: bridging the gap between client ideas and structured technical delivery. Clients use the portal to define their MVP ideas, run a collaborative agent swarm analysis, map out their visual roadmap, plan budgets, and monitor sprint updates in real-time.
Aegis is designed around clean architectural separation, with a fully asynchronous event-driven backend and a highly polished, responsive Single Page Application (SPA) frontend.
graph TD
subgraph "Client-Side (React & Vite)"
A[MVP Estimator UI] -->|Post Scope Description| B[Live Swarm Progress Monitor]
B -->|Server-Sent Events| C[Interactive Client Workspace]
C --> D[Roadmap SVG Timeline]
C --> E[Sprint Cost & Timeline Planner]
C --> F[Generated Tech Spec Blueprint]
end
subgraph "Server-Side (FastAPI & Python)"
G[FastAPI Server] -->|SSE Stream /api/mvps/stream| B
G -->|API Endpoints| H[Swarm Orchestration Engine]
H -->|Spawns Swarm| I[Product Manager Agent]
H -->|Spawns Swarm| J[System Architect Agent]
H -->|Spawns Swarm| K[Financial Analyst Agent]
I & J & K -->|Emit Logs & Updates| L[Thread-safe Mock DB]
L --> G
end
- Clean Layered Architecture (Python): Clear demarcation of API routers, domain logic (agent swarm and estimators), schemas (Pydantic), and database storage.
- Server-Sent Events (SSE): Asynchronous real-time log and state streaming from backend agents to client dashboard without HTTP polling.
- SVG Graph Visualization: Responsive SVG DAG mapping the MVP workflow phases (Ideation, Design, Development, QA, Launch) with active step indicator glows.
- Reactive Calculation State: Real-time frontend recalculations of team size, delivery date, and sprints as users adjust scope priorities or timeline velocity.
- DevOps Excellence: Dual-service containerization via Docker and Docker Compose, linting standards, and backend unit testing via pytest.
- Python 3.11+
- FastAPI: Fast, asynchronous web framework.
- Pydantic v2: High-performance data validation.
- Pytest: Unit testing and assertions.
- React 18 with TypeScript
- Vite: Ultra-fast build tool and bundler.
- Tailwind CSS: Contemporary styling system.
- Lucide Icons: Clean, scalable iconography.
- Docker & Docker Compose
- GitHub Actions: Continuous integration workflow.
Make sure you have Docker and Docker Compose installed on your machine.
- Clone the repository and navigate to the project root:
cd mvp-development - Build and start the services:
docker-compose up --build
- Open your browser and navigate to:
- Client Dashboard: http://localhost:5173
- API Swagger Documentation: http://localhost:8000/docs
- Navigate to the
backend/directory:cd backend - Create and activate a virtual environment:
python -m venv venv source venv/bin/activate - Install dependencies:
pip install -r requirements.txt
- Run the FastAPI development server:
uvicorn app.main:app --reload --port 8000
- Navigate to the
frontend/directory:cd ../frontend - Install dependencies:
npm install
- Run the Vite development server:
npm run dev
We value high test coverage and solid software contract assertions.
Navigate to the backend/ directory and run pytest:
pytestA client enters their raw product text. The backend spins up three concurrent virtual agents to collaborate on the specification:
- Product Manager: Synthesizes details into tangible user stories (Backlog items).
- System Architect: Designs database tables, key APIs, tech stack, and structure.
- Financial Analyst: Evaluates development sprint complexity, sizes team, and calculates budget.
The client watches the planning session unfold. As agents log thoughts and produce blueprints, Server-Sent Events stream details directly into the UI logs and timeline cards.
A beautiful visual representation of the stages to MVP launch. Completed stages display with checkmarks, active stages glow with animations, and pending stages stay blurred.
Clients can drag feature scopes, toggle priority levels, or speed up the delivery timeline with a slider. Costs, sprints, and launch date estimates recalculate instantly in the browser.
Distributed under the MIT License. See LICENSE for details.