License

This project is licensed under the MIT License.

Flowpilot is a one of a kind n8n workflow generator.Generation of production ready, deployable and testable workflows is what Flowpilot strives in.Flowpilot helps coders and none coders alike.Its designed to specifically understand your intent and translate it into a production ready workflow.

General Communication Flow

High-level data flow between the frontend, backend and database.

Copilot Communication Flow

n8n Generation workflow between the frontend, backend, generation module, LLM, vector-db and an open SSE connection.

ER Diagram

Main Features

AI Copilot

An intelligent workflow generation assistant that uses LLM technology to understand user intent and automatically generate production-ready n8n workflows.

• Real-time streaming workflow generation with progress tracking
• Multi-stage generation pipeline with user feedback
• Workflow history and chat management
• Trace and debugging support for generated workflows
• Server-sent events (SSE) connection for live updates

Community Hub

A collaborative platform for automation builders to share workflows, tips, and integrations with the n8n community.

• Infinite scroll feed of community posts and workflows
• Create and share posts with the community
• Browse and discover shared workflows by other users
• User profile cards with avatar and metadata
• Engage with other builders in the n8n ecosystem

User Profiles

Comprehensive user profile system with detailed information and engagement statistics across the platform.

• Customizable avatars and profile pictures
• Display user workflows and community posts
• View follower and following lists
• Track engagement metrics (likes, imports on workflows/posts)
• Manage profile settings and preferences
• Download workflow content and history

Landing

Login

Signup

Copilot

Simple workflow generation

Complicated workflow generation

Community Hub

Profile

Settings

This project was built using a test-driven and validation-first mindset. Every major feature was developed alongside tests to ensure reliability, prevent regressions, and maintain production-grade stability as the system evolved.

Frontend

Folder Structure	Centralized Error Handler	API Interceptor

Backend

Controller	Service	Centralized Error Handler

Github Actions

Git Philosophy

This project follows a structured Git workflow designed to keep the codebase clean, scalable, and production-ready at all times.

Branching Model

Branch	Purpose
main	Production-ready, always stable and deployable
dev	Integration branch where completed features are merged

No direct commits are made to main.

---

Branch Types

All work is created from dev using task-based branches:

Prefix	Usage
feature/	New features
fix/	Bug fixes
hotfix/	Urgent production fixes
refactor/	Code improvements
chore/	Tooling, config, dependencies
docs/	Documentation updates

Naming format: type/short-description

Examples:

feature/user-auth
fix/login-validation
refactor/api-layer
docs/readme-update

---

Tests

Feature tests

Unit tests

Deployment Architecture

FlowPilot uses Docker to containerize the frontend, backend, and database components, ensuring consistent deployment across development, staging, and production environments.

Database Configuration

Database setup and initialization using Docker:

Frontend Deployment

The frontend is containerized separately for both development and production environments:

Development Frontend	Root Docker Frontend Config

Backend Deployment

The backend services are containerized with appropriate configurations for both development and production:

Development Backend	Root Docker Backend Config

Development vs Production Separation

FlowPilot maintains a clear separation between development and production environments using Docker Compose configurations:

Development Environment

Command: docker compose -f docker-compose.dev.yml up --build

• Frontend: Uses Vite dev server with hot module reloading (HMR) for instant code feedback
• Backend: Runs in local mode with debugging enabled (APP_DEBUG=true)
• Volumes: Source code is mounted for live code synchronization without rebuilding
• Database: Uses ephemeral MySQL container (data not persisted)
• File Watching: Chokidar polling enabled for cross-platform file change detection
• Port Mapping: Frontend on 3000, Backend API on 8000

Production Environment

Command: docker compose -f docker-compose.prod.yml up --build -d

• Frontend: Multi-stage build producing optimized static files served by Nginx
• Backend: Production-optimized Laravel with caching and route optimization (APP_DEBUG=false)
• Volumes: Only configuration files are mounted; source code is copied during build
• Database: Named volume (db_data) for persistent data across container restarts
• Restart Policy: restart: always ensures services recover from failures
• Environment Variables: Loaded from .env.prod for sensitive configuration
• Detached Mode: Services run in background (with -d flag)

Key Architectural Differences

Aspect	Development	Production
Build Strategy	Lightweight, includes dev dependencies	Multi-stage, optimized for size and performance
Code Mounting	Live volumes for hot reload	Copied into image at build time
Persistence	Ephemeral (data lost on container stop)	Named volumes for database durability
Debugging	Enabled with verbose logging	Disabled for security and performance
Auto-Recovery	Manual restart required	Automatic via restart policy