MoonshotAI / MoonshotAI/kimi-code
Permissions handling are not working well: you start a plan mode and it auto-approve itself when in mode auto or yolo.
Nobody has claimed this yet.
- Dominant language
- TypeScript
- Stars
- 7.5k
- Forks
- 1.2k
- Avg merge
- 11h 53m
- Merged PRs (30d)
- 350
Description
What version of Kimi Code is running?
0.23.3
Which open platform/subscription were you using?
default
Which model were you using?
2.7 code
What platform is your computer?
Darwin 24.6.0 x86_64 i386
What issue are you seeing?
let's make a brainstorming/draft of the hybrid approach, replacing some of our skills and having xxx as a dependency
What steps can reproduce the bug?
It created the plan and started executing. There are many situations when I ask to plan, or manually set plan mode, and it ignores.
The root cause is the permission management, that since the new Kimi Code CLI was introduced, made "auto" and "yolo" very bad to use, while the manual mode is practically unusable, as it asks all the time for minor things.
Previous Kimi Code CLI was great, but this one is almost making me give up and move to another harness, after many bad behaviors that are making it unreliable.
What is the expected behavior?
Respect a plan mode.
Additional information
No response
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No files or tests are named in the report. Reproduce the behavior by comparing plan mode with auto, yolo, and manual permission modes, then trace the permission management and plan-mode entry points. Done means plan mode creates a plan without starting execution until the intended approval behavior occurs.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- authorization, cli
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100