Cost-Efficient Codex Subagents: Astra/Sol Orchestrator + Luna Workers

Melvin Vivas · X post · 2026-09-26 · Open on X

Topics: AI Agents, Tool Use & MCP, AI Dev Tools & Productivity, LLMOps, Deployment & Monitoring · Level: intermediate

Summary

How to set up subagents in Codex without using up usage limits too fast. Set each subagent's model and reasoning level explicitly. Use a strong model (Astra or Sol) to orchestrate and a lighter model (Luna) for the subagents. The creator's GitHub repo has a ready-made setup.

Key points

• Set each subagent's model and reasoning level explicitly instead of relying on defaults.
• Use Astra or Sol as the orchestrator model.
• Use Luna for the subagents that do the work.
• If you leave defaults in place, subagents burn through your usage limits fast.
• The donvito/codex-astra-luna-orchestrator repo has a ready-made version of this pattern.

Resources mentioned

• donvito/codex-astra-luna-orchestrator · repo · github.com · free
  The creator's Codex skill for an orchestrator-plus-subagents setup (Astra orchestrator, Luna subagents), configured through config.toml.
  Also in: Codex Orchestrator Pattern: Astra/Sol Orchestrator with Luna Subagents (Melvin Vivas on X · notes), Open-source Codex skill: Astra/Sol orchestrator with Luna subagents (Melvin Vivas on X · notes), Codex Orchestrator v0.2.1: GPT-6 Sol orchestrator with Luna subagents (Melvin Vivas on X · notes), Orchestrator/Subagent Patterns: Astra-Luna Skill vs Fusion and Advisor (Melvin Vivas on X · notes) and 27 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

Try this

• Set the model and reasoning level explicitly for every Codex subagent.
• Use Astra or Sol to orchestrate and Luna for subagents.
• Try the codex-astra-luna-orchestrator repo.