TheMoonMidas Shares Practical ChatGPT/Codex Tips for Fewer Misreads and Longer Quotas
On Sept 13, TheMoonMidas posted a thread of practical ChatGPT/Codex tips aimed at reducing intent misreads, cutting unnecessary permission interruptions, and stretching Codex quota further. The unifying idea: give the model concrete, pointable signals instead of vague requests. The closing advice is to pick just one tip that addresses your current pain point, try it once, and bake it into your instructions if it works.
Confirmed
- Provide references first: before asking ChatGPT for design work, share a few visual references and call out what you like in each (spacing, colors, navigation); he cites @BreejeAnadkat's full workflow.
- Be specific in feedback: avoid "make it better." For design, name concrete issues (e.g., weak headline, competing buttons, mobile menu covering content); for writing, paste the exact sentences you dislike and explain why (e.g., "reads like LinkedIn-speak; use everyday words, keep useful details, drop grand conclusions").
- Diagnose permission pauses: when Astra/Codex keeps stopping to ask permission, ask it to "show me the exact rule and file that made you pause" — often an old rule is the trigger.
- Set autonomy boundaries: use a reusable template — "build a runnable draft and verify locally; make reasonable choices on reversible details; ask me before publishing or spending" — reserving human sign-off for high-risk actions.
- Quota-saving orchestration: relaying @uzairansar's Codex Plus tip — run implementation on Luna max, set Astra as advisor, and have Luna call Astra only when stuck or looping, using Codex's built-in agent-to-agent tooling.
- Compare costs with logs: hand your usage log to ChatGPT with a prompt like "compare usage, time and finished work across these runs" to evaluate orchestration options.
Why it matters
The shared methodology — clear stopping points, pointable feedback, and a clear division of labor (implementer vs. advisor) — reduces rework and directly lowers token consumption, especially valuable for quota-limited Codex Plus users.
2026-09-13 ~ 2026-09-13 · 9 related posts
Primary sources
- [source] A practical tips thread for fixing Codex misfires, pause loops, and token burn — TheMoonMidas · 2026-09-13
- [source] Show ChatGPT visual references before design work, and say what you like in each — TheMoonMidas · 2026-09-13
- Vague "make it better" fails: point at specific design problems instead — TheMoonMidas · 2026-09-13
- Stale rules in agents.md may cause Codex's Astra to pause constantly — ask it to show you — TheMoonMidas · 2026-09-13
- Give Codex a clear stopping point: reversible details on its own, ask before publishing or spending — TheMoonMidas · 2026-09-13
- [source] Codex Plus tip: run Luna for implementation and call Astra as an advisor when stuck — TheMoonMidas · 2026-09-13
- Feeding your usage log to ChatGPT to compare Codex orchestrator setups on cost — TheMoonMidas · 2026-09-13
- Show ChatGPT the exact sentence you hate instead of vague rewrite requests — TheMoonMidas · 2026-09-13
- Adopt one tip at a time: fix your current annoyance, then bake it into your instructions — TheMoonMidas · 2026-09-13