# Custom subagent creation in codex

**URL:** <https://community.openai.com/t/custom-subagent-creation-in-codex/1388744>\
**Category:** ChatGPT Apps SDK\
**Tags:** chatgpt\
**Created:** [August 2, 2026, 4:32am UTC](https://community.openai.com/t/custom-subagent-creation-in-codex/1388744 "2026-08-02T04:32:44Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![gemini.ye100](https://avatars.discourse-cdn.com/v4/letter/g/b77776/32.png) [@gemini.ye100](https://community.openai.com/u/gemini.ye100)\
**Post date:** [August 2, 2026, 4:32am UTC](https://community.openai.com/t/custom-subagent-creation-in-codex/1388744/1 "2026-08-02T04:32:44Z")

</div>

Hi team,

I’m trying to create a non-forked custom Codex subagent using the documented configuration approach.

I created this file at `~/.codex/agents/luna_high.toml`:

```auto
name = "luna_high"
description = "GPT-5.6 Luna agent for narrowly scoped, repeatable work that needs high reasoning effort."
model = "gpt-5.6-luna"
model_reasoning_effort = "high"

developer_instructions = """
Handle only the bounded task delegated by the parent agent.
Work independently, keep the scope narrow, and return a concise evidence-backed summary.
Do not delegate further work unless the parent explicitly asks.
"""

```

The Codex documentation says custom agents can define `model` and `model_reasoning_effort`, and that Codex identifies them by their `name`.

However, when I try to spawn this custom agent from the desktop app’s subagent workflow, the runtime rejects it with:

```auto
unknown agent_type 'luna_high'

```

The runtime currently appears to expose only built-in agent types such as `default`, `worker`, and `explorer`, and model overrides appear limited to `gpt-5.6-sol` and `gpt-5.6-terra`.

Could you clarify:

1. Is `gpt-5.6-luna` currently supported for custom subagents in Codex desktop?
2. Are custom agents from `~/.codex/agents/*.toml` expected to be selectable in the desktop app’s subagent workflow?
3. Does adding or changing a custom agent require restarting Codex or creating a new task?
4. If the UI/runtime does not support custom agent names yet, what is the supported way to spawn a non-forked Luna High subagent?
5. Is there an official runtime/status view that confirms the actual model and reasoning effort used by a spawned subagent, rather than only showing its requested configuration?

Environment: Codex desktop app, Codex CLI `0.146.0`, ChatGPT sign-in.

---

<div class="post-metadata">

**Author:** ![galdx.adm](https://avatars.discourse-cdn.com/v4/letter/g/3be4f8/32.png) [@galdx.adm](https://community.openai.com/u/galdx.adm)\
**Post date:** [August 4, 2026, 3:07am UTC](https://community.openai.com/t/custom-subagent-creation-in-codex/1388744/2 "2026-08-04T03:07:51Z")

</div>

had the same issue now. i was able to make luna available for subagents only when i went the path settings \> workspace dependencies \> reset and install workspace. in a new session, i was able to create luna agents, but in old ones it is still stuck with only terra and sol.

it was likely a bug caused in one of the updates.

---

<div class="post-metadata">

**Author:** ![OpenAI\_Support](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/openai_support/32/561845_2.png) [@OpenAI\_Support](https://community.openai.com/u/OpenAI_Support)\
**Post date:** [October 9, 2026, 7:03pm UTC](https://community.openai.com/t/custom-subagent-creation-in-codex/1388744/3 "2026-10-09T19:03:05Z")

</div>

Hey @gemini.ye100 and @galdx.adm,

Sorry for the delayed response.

GPT-5.6 Luna is one of the supported models for subagents, and you should be able to assign names to custom subagents.

Can either of you confirm whether you're still encountering this issue with subagents?

-Luke