Optimized ChatGPT Development Process for Code Projects

I’ve been testing different approaches with AI coding assistants and found a method that really helps maintain project structure. When I need the AI to build new functionality or improve existing code, I use this framework:

:magnifying_glass_tilted_left: PHASE 1: UNDERSTAND PROJECT SCOPE
AI, review the guidelines in my project-rules.md file, then think through this systematically before moving forward.
PAUSE. Confirm you grasp these key points:

  1. This focuses on code optimization and reusability
  2. New file creation needs strong reasoning
  3. All recommendations must link to current codebase
  4. Breaking these guidelines makes responses unusable

BACKGROUND: Last contractor ignored existing structure and made duplicates everywhere. Need to show you can work with what’s already built.

REQUIRED WORKFLOW:

  1. Begin with “UNDERSTOOD: I will focus on extending existing code”
  2. Review current files BEFORE proposing changes
  3. Point to actual files from your review
  4. Add verification steps in your response
  5. Close with confirmation of approach

GUIDELINES (breaking these invalidates everything):
:cross_mark: No new files without thorough justification
:cross_mark: No complete rewrites when updates work
:cross_mark: No vague suggestions - give concrete solutions
:cross_mark: No bypassing current architecture
:white_check_mark: Build on existing modules
:white_check_mark: Merge redundant code
:white_check_mark: Show exact file locations
:white_check_mark: Include transition plans

:wrench: PHASE 2: EXAMINE EXISTING CODE
Look at current project structure and find files related to what I’m requesting.

:clipboard: PHASE 3: BUILD ACTION PLAN
Using your findings, make a clear plan for implementing the feature.

:gear: PHASE 4: TECHNICAL IMPLEMENTATION
Show specific code modifications, API updates, and how everything connects.

:white_check_mark: PHASE 5: WRAP UP DELIVERABLES
Finish with testing approach, deployment notes, and final suggestions.

This structured approach stopped the AI from inventing files or breaking my organization. It behaves more like a careful teammate instead of creating chaos.

My project-rules.md contains:

  • Don’t generate files that don’t exist
  • Don’t invent features not in my actual project
  • Don’t bypass my current setup
  • Only modify existing files and structure
  • Stay accurate and respect the codebase

I also get another AI to examine my codebase first, then review the implementation plan afterward. This double-check process works great for adding features. Anyone have other ideas to improve this further?

thats solid advice! i’d add one more thing - have the AI summarize your project-rules.md in its own words before it starts coding. they’ll say “got it” but still screw up basic requirements. also ask for an impact estimate on existing functionality first. catches problems before they happen.