MoonshotAI / MoonshotAI/kimi-cli

subagent 无线循环 || subagent wireless loop

Open
#1,927 5 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug
Dominant language
Python
Stars
11.4k
Forks
1.3k
Avg merge
9h 47m
Merged PRs (30d)
2

Description

What version of Kimi Code CLI is running?

1.36.0

Which open platform/subscription were you using?

/login

Which model were you using?

kimi-for-coding

What platform is your computer?

Darwin 25.4.0 arm64 arm

What issue are you seeing?

启动 subagent 执行任务,会反复的读同一个文件上百次,无线循环,持续了半小时

Image
What steps can reproduce the bug?

我有 2 个大任务,每个大任务都是root agent提方案并拆解 -> boundary subagent-> test subagent -> dev subagent -> root agent 这样的 agent 流水线,第一个大任务已经完成了,进入第二个大任务的 test 时出的问题。

What is the expected behavior?

No response

Additional information

No response


What version of Kimi Code CLI is running?

1.36.0

Which open platform/subscription were you using?

/login

Which model were you using?

kimi-for-coding

What platform is your computer?

Darwin 25.4.0 arm64 arm

What issue are you seeing?

When the subagent is started to perform a task, the same file will be read hundreds of times in a wireless loop that lasts for half an hour.

Image
What steps can reproduce the bug?

I have 2 big tasks. Each big task is an agent pipeline like root agent proposing and dismantling -> boundary subagent-> test subagent -> dev subagent -> root agent. The first big task has been completed, but there is a problem when entering the test of the second big task.

What is the expected behavior?

No response

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

Start by reproducing the loop with Kimi Code CLI 1.36.0 using the described root, boundary, test, and dev subagent pipeline, then inspect the trace showing repeated reads of the same file. Done means the second task's test subagent completes without repeatedly reading one file or running indefinitely.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
ai, cli
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.