MoonshotAI / MoonshotAI/kimi-code
"Approve once" and "Approve for this session" behave identically — no session-level memory
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?
0.9.0
Which open platform/subscription were you using?
ubuntu
Which model were you using?
kimi-k2.6
What platform is your computer?
linux
What issue are you seeing?
In the latest kimi-code (v0.9.0), selecting "Approve once" vs "Approve for this session" when a tool asks for permission results in identical behavior — the next tool call in the same session still prompts for approval. The session-level approval appears to be broken or not implemented.
What steps can reproduce the bug?
In the latest kimi-code (v0.9.0), selecting "Approve once" vs "Approve for this session" when a tool asks for permission results in identical behavior — the next tool call in the same session still prompts for approval. The session-level approval appears to be broken or not implemented.
What is the expected behavior?
In the latest kimi-code (v0.9.0), selecting "Approve once" vs "Approve for this session" when a tool asks for permission results in identical behavior — the next tool call in the same session still prompts for approval. The session-level approval appears to be broken or not implemented.
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
Reproduce the permission prompt behavior in Kimi Code v0.9.0, then trace the tool-approval and session state handling. Done means “Approve for this session” suppresses later prompts in that session while “Approve once” continues to apply only to the current tool call.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- authorization, cli
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100