Communicate the human readable subagent name when spawning

Resolved 💬 3 comments Opened Feb 19, 2026 by smileBeda Closed Feb 27, 2026
💡 Likely answer: A maintainer (github-actions[bot], contributor) responded on this thread — see the highlighted reply below.

What variant of Codex are you using?

CLI

What feature would you like to see?

Currently when the CLI tool is spawning a subagent it communicates its internal ID, which is fairly useless to the human, see below as an example:

• Agent spawned
  └ call: call_sWG3YeF4FYsJR6aD1BLRpPV4
    agent: 019c76a2-03e2-7c51-b307-80d8ba2c450f
    status: pending init
    prompt: Inspect /CWD/ and report: existing files, languages/frameworks, and any constraints from repository docs beyond AGENTS.md.
Provid...

It is possible to me in this case to _guess_ that it is my scouting agent, but - I have 12 agents registered, and several of them do scanning work.

As such, this type of message would be MUCH more useful:

• Agent spawned
  └ call: call_sWG3YeF4FYsJR6aD1BLRpPV4
    agent: scout (the name of the agent from `[agents.scout]`)
    status: pending init
    prompt: Inspect /CWD/ and report: existing files, languages/frameworks, and any constraints from repository docs beyond AGENTS.md.
Provid...

Additional information

This would help a lot to see what agents it actually spawns, not only for control but also to eventually force other agents, and so on.

View original on GitHub ↗

3 Comments

github-actions[bot] contributor · 6 months ago

Potential duplicates detected. Please review them and close your issue if it is a duplicate.

  • ##12047

Powered by Codex Action

smileBeda · 6 months ago

That is not a duplicate, more of a "catch all what I'd like in one big ticket and ..."

This here is a concise and implementable/actionable request instead

smileBeda · 6 months ago

This is resolved in 105