Use Plan mode for underspecified or risky tasks
Engage Plan mode when tasks are underspecified, risky, or touch multiple systems.
Plan mode asks Codex to propose the implementation path before changing code.
Before making any code changes, using Plan mode prompts the AI to outline its proposed approach. This is particularly useful for complex, uncertain, or multi-system tasks, allowing for review and adjustment of the strategy before execution.
Before asking Codex to refactor a large module, use Plan mode to review its proposed steps.
Utilize Plan mode to have the AI outline its implementation strategy before altering code. This is beneficial for tasks that are underspecified, carry risk, or involve multiple interconnected systems.
Allows for a strategic review of the proposed approach, ensuring alignment and mitigating risks before any code is modified.
- Mastering remote engineering work from your phone ↗OpenAI Developers Blog · OpenAI Developers Blog · Sep 3, 2026 · paraphrased
First seen on Subvolts Sep 3, 2026. Updated Sep 11, 2026. Spotted a problem? Tell us.
- Mastering remote engineering work from your phoneOpenAI · OpenAI Developers Blog · Sep 3, 2026
The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.
- PromptGenerate a complex 3D simulation of a town's evolutionBuild a procedurally generated 3D simulation of a harbor town's evolution from 3000 BCE to 3000 AD. · 30 min
- HeuristicGPT-5.5 shows stronger reasoning and autonomyLeverage GPT-5.5 for its enhanced reasoning and autonomy in complex software engineering tasks. · 3 min
- ComparisonGPT-6.1 Sol vs. Astra for cost-performance tradeoffCompare GPT-6.1 Sol and Astra on your tasks to find the best balance of quality and cost. · 10 min
- Cost tipOptimize GPT-6.1 Sol token costsUnderstand GPT-6.1 Sol token pricing for input, cache, and output to manage costs effectively. · 5 min
- Setting tipGPT-6.1 Sol data residency optionsChoose between US and EU data residency for GPT-6.1 Sol, noting potential feature limitations. · 1 min
- Setting tipGPT-6.1 Sol reasoning effort levelsSet the reasoning effort for GPT-6.1 Sol to control depth and cost. · 1 min
Questions people ask
- Which AI tools does this work with?
- The sources used it with ChatGPT.
- Where did this come from?
- OpenAI Developers Blog (OpenAI Developers Blog). Subvolts collected, normalized and credited it; the link above opens the original.