ChatGPT standard “Repo Command Set”
March 18th, 2026 1:57 PM Mr. Q Categories: Programming
ChatGPT + GitHub Connector Command System (Ultra + Stateful Combined)
Use this to review and navigate any repo without PRs or copy/paste.
⚡ One-Line Super Command (Fastest)
Use the GitHub connector to fully review ToolboxAid/HTML-JavaScript-Gaming.
map the repo, identify key entry points and engine files, review the last 5 commits, analyze architecture, scan for bugs and performance issues, focus on physics, collision, object lifecycle, and game loop timing
Then:
- produce a prioritized action plan
- show quick wins, medium refactors, and major upgrades
⚡ MASTER COMMAND (Full Audit + Action Plan)
Use the GitHub connector to fully analyze ToolboxAid/HTML-JavaScript-Gaming.
Perform:
1. Repo map (structure + key files)
2. Architecture review
3. Review last 5 commits
4. Identify bugs and risks
5. Identify performance issues
6. Suggest top 5 improvements
Focus on:
- physics correctness
- collision handling
- object lifecycle
- game loop timing
Then:
- provide a prioritized action plan
- separate quick wins vs major refactors
🧠 Start Session (Do This First in a New Chat)
Use the GitHub connector to start a persistent review session for ToolboxAid/HTML-JavaScript-Gaming.
Do all of the following:
- map the repository
- identify key entry points, engine files, and major systems
- identify physics, collision, rendering, input, and object lifecycle systems
- review the latest commits
- identify current risks, weaknesses, and opportunities
Then create and maintain a review state with:
- Repo Map
- Current Priorities
- Open Issues
- Suggested Refactors
- Performance Risks
- Next Best Actions
Then:
- keep this review state active for the entire chat
- treat all future prompts as updates to this state
🧠 Stateful Review System (Persistent Intelligence)
Session Continue Command
Use the GitHub connector to continue the current review session.
Update:
- Current Priorities
- Open Issues
- Suggested Refactors
- Performance Risks
- Next Best Actions
Then:
- tell me the single best thing to work on next
Commit Tracking Command
Use the GitHub connector to continue the session and review latest commits.
For each relevant commit:
- summarize the change
- determine whether it improves or worsens the current architecture
- identify bugs, regressions, or performance risks
- update the review state accordingly
Then:
- highlight regressions and risky changes
Area-Focused Command
Use the GitHub connector to focus on a specific system or folder.
Within <system or folder>:
- explain current design
- identify issues and risks
- suggest fixes
- update the review state
- classify as quick win, medium refactor, or major rebuild
Then:
- recommend fix vs refactor vs rewrite
Fix Planning Command
Use the GitHub connector with the current review state.
Turn the top issues into a practical fix plan.
For each item include:
- problem
- why it matters
- files likely involved
- recommended fix
- risk level
- implementation priority
Then:
- order fixes from highest to lowest impact
Rewrite Planning Command
Use the GitHub connector with the current review state.
Propose a rewrite plan for the weakest subsystem.
Include:
- current weakness
- target design
- files to replace or refactor
- migration steps
- risks
- expected benefit
Then:
- decide if rewrite should happen now or later
💾 How to Save Your Plan (Critical for Workflow)
Option 1 — Save in Chat (Fastest)
Save this as the current review state and treat it as the source of truth for future updates
Option 2 — Export to File
Convert the current review state into a clean markdown document I can save locally
Option 3 — Turn Into Task List
Convert the current review state into a prioritized task list with checkboxes
Option 4 — WordPress Post Format
Convert the current review state into a WordPress Classic Editor post with sections, callouts, and a summary
Option 5 — Ongoing Tracking
Track completed vs remaining items and update the review state as tasks are finished
🔧 Core Command Set (Modular Use)
🧭 Repository Map / Navigator
Use the GitHub connector to map the repository.
Show:
- folder structure (high level)
- key entry points
- core engine files
- physics / collision related files
- object lifecycle classes
Then:
- suggest the best review order
⚙️ Architecture Review
Use the GitHub connector to analyze architecture.
Focus on:
- separation of concerns
- physics engine design
- rendering vs logic separation
- scalability
- suitability for robotics integration
Then:
- provide a refactor plan
🔍 Latest Commits Review
Use the GitHub connector to review recent commits.
For each commit:
- summarize changes
- identify bugs or risks
- suggest improvements
- call out performance issues
Then:
- highlight the most important changes
🔬 Deep System Review
Use the GitHub connector to compare recent changes.
Focus on:
- physics correctness
- collision handling
- object lifecycle
- game loop timing
Provide structured issues and fixes
Then:
- prioritize critical fixes
📁 File-Level Review
Use the GitHub connector to review a file.
Explain:
- purpose
- bugs
- performance issues
- refactoring suggestions
Then:
- recommend improvements
Example:
Review src/physics.js
🧩 Folder / System Review
Use the GitHub connector to analyze a folder.
Explain:
- architecture
- responsibilities
- weaknesses
- improvements
Then:
- identify weakest area
🐞 Bug Hunt Mode
Use the GitHub connector to scan for bugs.
Focus on:
- collision edge cases
- NaN / undefined risks
- memory leaks
- timing issues
Then:
- list likely bugs
⚡ Performance Review
Use the GitHub connector to analyze performance.
Focus on:
- loops
- object allocations
- physics calculations
- rendering inefficiencies
Then:
- identify biggest bottlenecks
🧭 What Should I Work On Next?
Use the GitHub connector to analyze priorities.
Provide:
- top 5 improvements
- highest risk areas
- best next refactors
- quick wins vs major upgrades
Then:
- recommend next task
🔁 Daily Workflow Command
Use the GitHub connector to continue your review session.
Map changes, review recent commits, analyze architecture, track bugs and performance risks, focus on physics, collision, object lifecycle, and game loop timing, update the review state
Then:
- tell me what to fix next
🚀 Advanced Power Commands
Drill Down
Go deeper into collision handling and show exact code fixes
Force Rewrite
Rewrite this system using best practices and improved performance
Prioritize Next Work
Based on this repo, what should I fix next for maximum impact
🧩 Workflow Strategy
First Time (New Chat)
- Run Start Session
Daily Use
- Run Daily Workflow Command
Debugging
- Use Deep System Review or Bug Hunt
Refactoring
- Use Architecture Review or Rewrite Plan
Focused Work
- Use File-Level or Folder Review
🏷️ Work Classification System
- Quick Wins → small, high-impact fixes
- Medium Refactors → moderate effort
- Major Rebuilds → redesign required
- Watch List → risky areas
💡 Why This Works
- No PRs required
- No copy/paste needed
- Works with commit workflow
- Built for game engine + physics systems
- Extends into robot architecture systems