Helper agentssubagents
Luna Max as a Codex subagent
Bounded delegated work at max reasoning: a luna_worker custom agent, checked against your Codex version, diff shown first.
6.1K saves on X, as of Posted
Prompt
create a custom agent named luna_worker at ~/.codex/agents/luna-worker.toml.
use these settings:
model = "gpt-5.6-luna"
model_reasoning_effort = "max"
give it a description and instructions for bounded delegated work. preserve the rest of my config. validate it against my installed Codex version, show me the diff, then use luna_worker for subagent tasks.Next Send it to Sol in Codex.
Tell Sol to configure Luna Max as a subagent. The prompt creates a custom agent file, keeps the rest of your config, validates it against your installed Codex version and shows you the diff before using it.
Another option: skip the custom agent file. Keep Sol in the main thread and ask it to spawn separate Luna Max agent threads for each task. In Codex, a thread is a separate conversation with its own context. Several can run at once, and Sol collects their results in the main thread.
Share this tip
