June 3, 2025 · essay
From CLI to Electron GUI - A SDLC Workflow Reimagined
Discover how the Python 'ai-sdlc' CLI was transformed into SDLC a user-friendly graphical module in our Electron desktop app, by Google Gemini in under a days work. A look at AI-driven re-implementation and future integrations.
The short version
- LLM
- Google Gemini 2.5 Pro Preview, guided by a detailed System Instruction and existing Python code as specification.
- Why
- To rapidly re-implement the Python-based 'ai-sdlc' CLI tool as SDLC a native Electron/Vue 3 GUI module, making a structured SDLC process more accessible and visually manageable within an existing AI toolkit application.
- Challenge
- Translating Python CLI logic into TypeScript Electron services, building a comprehensive Vue 3 UI for the 8-step SDLC (Software Development Life Cycle) workflow, and ensuring seamless integration within the desktop app—all orchestrated through AI-assisted development in under 6 hours.
- Outcome
- A fully functional SDLC module integrated into the desktop application. It provides GUI-based project initialization, feature management, step navigation, AI prompt generation, and workflow advancement, effectively mirroring and enhancing the original CLI's capabilities.
- AI approach
- An AI-First methodology where Gemini 2.5 Pro performed 100% of the code generation (TypeScript services, Vue 3 UI) for SDLC. The human developer acted as architect, prompter (using Python code as spec), and validator, guiding the AI to re-implement and integrate the SDLC functionality.
- Learnings
- Leveraging an existing codebase (like the ai-sdlc Python tool via code2prompt) as a clear specification for an AI coding assistant (Gemini) dramatically accelerates re-platforming and feature development. A strong System Instruction and modular design are key to managing AI-driven projects of this complexity and achieving high-quality results quickly.
From CLI to GUI SDLC Reimagines an SDLC Workflow
Today marked a big step in enhancing our desktop application toolkit. We focused on integrating a sophisticated new module, now dubbed SDLC designed to bring a graphical user interface to a structured Software Development Life Cycle (SDLC) process. This project involved re-implementing the core logic of the open-source ai-sdlc Python CLI tool as a native TypeScript module within our existing Electron and Vue 3 desktop application. The entire process, from understanding the Python tool to having a functional GUI, was completed in a short timeframe – a days work – showcasing a highly efficient development approach, a core tenet of my AI-First Development Philosophy.
The Goal: A User-Friendly Interface for a Proven SDLC Workflow
The ai-sdlc CLI provides a robust 8-step methodology for guiding software development, particularly when leveraging AI assistance. While effective, a command-line interface can be less intuitive for some users. SDLC aims to make this structured process more accessible and visually manageable by:
- Re-implementing the ai-sdlc's file and state management logic (project initialization, feature tracking, step progression, prompt template handling) natively in TypeScript for our Electron application's main process.
- Creating a Vue 3 frontend that allows users to visually initiate projects, manage features, navigate SDLC steps, view generated AI prompts, and track progress.
- Integrating this module into our existing desktop application, allowing for future synergies with its other AI-powered features detailed in posts like The Anatomy of a Desktop AI Assistant.
The Development Process: A Rapid Transformation
Our development strategy centered on leveraging advanced tools and a clear plan, reflecting the principles discussed in 15 Power Tips for AI-First Development in AI Studio:
1. Understanding the Blueprint with code2prompt:
To ensure an accurate re-implementation, we first needed to provide our development environment (and our AI coding assistant, Google Gemini) with the complete context of the original Python ai-sdlc tool. We used code2prompt to consolidate the entire Python project into a single Markdown file. For those dealing with large codebases, I even built a Tkinter GUI helper for code2prompt to simplify this process. This consolidated output served as a precise technical specification.
2. Systematic Re-implementation (Python to TypeScript):
Guided by a detailed System Instruction defining the target architecture (Electron, Vue 3, TypeScript, Node.js file system operations, TOML/JSON parsing) and modular patterns—part of My Evolving AI Power Stack—our AI assistant, Gemini, systematically translated the Python logic. This involved creating utility functions, main process services mirroring each ai-sdlc command, and establishing the IPC/Preload layer.
3. Building the Vue 3 GUI:
With the backend logic in place, Gemini then constructed the user interface, including a dashboard for project management and a detailed workflow view for navigating the 8 SDLC steps, complete with prompt display and action buttons. This approach aligns with the idea that Conversational AI (could be) the New IDE.
4. Collaborative Debugging and Refinement:
Throughout the process, minor integration issues or Vue-specific warnings were identified and quickly resolved through an iterative feedback loop with Gemini, ensuring a polished final output.
The Result: SDLC – A Powerful SDLC Module
In less than six hours, the core of the ai-sdlc workflow was successfully transformed into SDLC a fully integrated module. Users can now graphically manage the entire lifecycle, from initializing projects to completing features, all within a cohesive desktop environment. This rapid iteration is a hallmark of truly embracing an AI's Near-Flawless Execution when guided properly.
Future Synergies and the Agnostic Approach
The SDLC module is designed to be LLM-agnostic at its core, just like the original ai-sdlc tool. It generates prompts that users can take to any AI coding editor, VS Code fork, AI-assisted IDE, or standalone AI chat interface. This flexibility, allowing users to choose from a variety of models much like using OpenRouter for its LLM variety, is key.
However, the real power will come from its integration with the other existing features of our desktop application. Imagine a "one-click" SDLC planning scenario where a feature request is automatically processed by a local Ollama model (powered by NVIDIA GPUs + Ollama for local AI) to flesh out an initial idea, with the output seamlessly becoming the input for SDLC's first step. When SDLC generates a prompt for the PRD, a button could "Send to Local AI Chat," pre-populating the prompt in our app's Ollama chat interface. The AI's response could then be saved directly to the correct SDLC Markdown file.
Similar integrations are envisioned for using the app's Vision feature to analyze UI mockups for PRD generation, leveraging the RAG system (which I've discussed in posts like To RAG or Not to RAG? and integrating Crawl4AI for RAG) to query existing project documents, and using Live Session summaries as input for new ideas.
Key Takeaways from this Rapid Integration:
- Providing a complete, existing codebase as a specification (via tools like code2prompt) is highly effective for AI-driven re-implementation.
- A clear System Instruction defining the target architecture and standards is crucial for AI coding assistants. Understanding how to effectively prompt is one of the 6 Essential Resources for AI Consultants & Developers.
- Modular design facilitates rapid, AI-assisted development of complex features. The importance of clear User Stories cannot be overstated here.
- The developer's role increasingly becomes one of high-level architecture, precise requirement definition, and meticulous validation, enabling extraordinary development velocity. This isn't about using AI as a simple AI hammer for throwaway apps, but for building robust systems.
The SDLC module is now poised for thorough end-to-end testing, UI refinements, and then the exciting phase of deep integration with the application's other AI tools. This project continues to solidify the practical power of building sophisticated tools with an AI-First approach, much like how AI can enhance workflows in Google Sheets & Apps Script or through platforms like n8n.