openai / openai/codex

When an agent uses subagents to implement code the files changes / lines written aren't counted correctly in the end

Open
#39,313 0 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

bug extension subagent
Dominant language
Rust
Stars
125k
Forks
19.5k
PR merge metrics
PR metrics pending

Description

What version of the IDE extension are you using?

26.814.41407

What subscription do you have?

ChatGPT Pro 20x

Which IDE are you using?

VS Code

What platform is your computer?

Darwin 25.5.0 arm64 arm

What issue are you seeing?

When an agent uses subagents to implement code the files changes / lines written aren't counted correctly in the end

Image
What steps can reproduce the bug?

Think of some random project / feature and get 5.6 Sol Ultra to build it and delegate work to subagents. When it finishes you will see that the lines reported in the end are only the ones the central agent wrote, not the subagents, thereby creating a misleading picture of the size of the implementation.

What is the expected behavior?

The expected behavior is to see all the lines changed by the agent AND the subagents combined, not just the central agent. It should equal the diff with what it had before the prompt was sent.

Additional information

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Reproduce the issue in the VS Code extension by asking an agent to delegate implementation to subagents, then compare the reported file and line counts with the final diff against the pre-prompt state. Trace the completion summary or usage-reporting entry point that produces those counts; done means the central agent and all subagent changes are included.

Written by the indexing model from the issue text.

Assessment

Tech stack
rust, vscode
Domain
developer-experience, tooling
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
55/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.