How to Prompt GPT-6 Astra in Codex: Simpler Instructions, Leaner AGENTS.md

Melvin Vivas · X post · 2026-09-12 · Open on X

Topics: Prompt & Context Engineering, AI Dev Tools & Productivity, AI Agents, Tool Use & MCP · Level: intermediate

Summary

A short checklist for adapting Codex setups to GPT-6 Astra. Because the model is more capable, prompts, skills and AGENTS.md files should get shorter and less prescriptive. They should still set clear decision boundaries and say clearly what 'done' means.

Key points

• Give the model fewer instructions; smarter models need less hand-holding
• Keep skills short
• Keep AGENTS.md lean
• Set clear decision boundaries (what the agent may decide on its own)
• Write a clear definition of done for each task
• Force less step-by-step process on the model

Resources mentioned

• GPT-6 Astra · tool · openai.com · paid
  The model announced in the quoted launch post, pitched as the developer's most capable model for work, coding, science and cybersecurity, and able to operate a computer.
  Also in: Customizing Your Coding Setup with Pi Coding Agent Extensions (Melvin Vivas on X · notes), Pi Agent Council: Ask Multiple LLMs in Parallel and Compare Their Advice (Melvin Vivas on X · notes), Use GPT-6.1 Sol by Default, Save Astra for Emergencies (Melvin Vivas on X · notes), Dots in ChatGPT: always-on AI agents that you hand responsibilities to (Melvin Vivas on X · notes) and 49 more
• OpenAI Codex · tool · openai.com · paid
  OpenAI's coding agent. In the diagram it writes code, fixes review findings and drives the build loop. The creator also used it to make this video.
  Also in: An agent bot that installs and drives Codex on its own (Melvin Vivas on X · notes), Asking a Coder bot to install Codex (Melvin Vivas on X · notes), Sign in with ChatGPT: Setting Usage Limits for Each App (Melvin Vivas on X · notes), Codex Cloud Environments Must Be Saved & Published Before Use (Melvin Vivas on X · notes) and 240 more
• AGENTS.md · docs · agents.md · free
  Instruction file format that gives coding agents project context such as the tech stack, skills and MCP servers.
  Also in: Claude Code Now Reads AGENTS.md When CLAUDE.md Is Missing (Melvin Vivas on X · notes), Subagents with mixed models: a strong planner and a fast executor (Melvin Vivas on X · notes), Updating Skills and AGENTS.md for Astra with OpenAI's Guidance (Melvin Vivas on X · notes), OpenAI Guide: Rethinking Skills and Prompts for GPT-6 Astra (Melvin Vivas on X · notes) and 5 more

Try this

• Cut down your AGENTS.md and skills files
• Add a clear definition of done and clear decision boundaries to task prompts
• Remove forced process steps and extra hand-holding