@Saccc_c: How to give your Codex unlimited bullets? Set the subagent to luna-max. Use sol for complex task design and luna-max for parallel execution. Specific setup: Enter the following prompt in the Codex conversation: In ~/.codex/agents/luna-worke…
Summary
Shares a method for customizing a Codex subagent (luna-max) and configuring it as luna-worker, allowing complex tasks to use sol for design and luna-max for parallel execution, resulting in stronger execution capability.
View Cached Full Text
Cached at: 08/03/26, 03:34 AM
How can you give your Codex unlimited bullets?
Set the sub-agent to luna-max. Use sol for solution design on complex tasks, and luna-max for parallel execution.
Specific setup method:
Enter the following prompt in the Codex conversation:
Create a custom agent named luna_worker in ~/.codex/agents/luna-worker.toml.
Use the following settings:
model = “gpt-5.6-luna”
model_reasoning_effort = “max”
Similar Articles
@Saccc_c: Someone has open-sourced the Codex infinite ammo method. The core remains that Sol is responsible for orchestration and planning, while Luna max handles the actual execution. This method saves about 59% of token cost on simple tasks, and reaches 65% on complex tasks.
Someone has open-sourced the Codex infinite ammo method, using Sol for planning and Luna Max for execution, saving about 59% of token costs on simple tasks and 65% on complex tasks.
@yangyanglxy: Codex Plus with Infinite Ammo is Here — The core of this design: Sol handles judgment, Luna handles throughput, and Sol handles fallback. Below is the complete configuration tutorial.
This article explains how to give Codex a Plus version with infinite ammo through custom sub-agent configuration: Sol handles planning and fallback, while Luna handles parallel execution. Includes a complete configuration tutorial.
@Lonely__MH: Great news for Codex users! When I opened Codex this morning and saw the reminder about using subagents, I happened to come across @Saccc_c's tweet. I gave it a full try — the approach is completely feasible! However, the prompt for creating subagents still needs improvement; you must add the corresponding description and working method. Below is my optimized complete prompt...
Explains how to create the custom subagent luna_worker in Codex, including the full prompt and configuration method, referencing @Saccc_c's approach to help users run tasks in parallel.
@Luckyjudy666: 8 Tips to Make Codex Your Personal Assistant 1. Build a Shared Memory for Codex Core rules go in Agents.md, project background in Obsidian, repeated processes as skills, personal preferences and common questions in Memories. Otherwise, Codex is like a new colleague every time, having to explain everything from scratch...
This article shares 8 tips for using the Codex AI assistant effectively, including building shared memory, remote task execution, scheduled automation, file organization, and teaching new software operations, all aimed at improving work efficiency.
@zjp1997720: Strongly recommend, a must-install plugin for Codex! A major issue I've had with Codex is that even for very complex tasks, it doesn't actively call Agents Team. In contrast, Claude Code is very proactive. I recently discovered that oh my Opencode (…
Recommends a plugin called Lazy Codex, which helps Codex actively call Agents Team to handle complex tasks, along with an installation link.