MoonshotAI / MoonshotAI/kimi-code
Feat: allow permanently skipping the /goal start permission prompt ("don't ask again")
Nobody has claimed this yet.
- Dominant language
- TypeScript
- Stars
- 7.5k
- Forks
- 1.2k
- Avg merge
- 11h 53m
- Merged PRs (30d)
- 350
Description
Feature request
When starting a goal with /goal <objective> in yolo (or manual) permission mode, the TUI always shows the "Start a goal in YOLO mode?" confirmation dialog (options: Switch to Auto and start / Keep YOLO and start / Do not start).
There is currently no way to opt out: no config key in config.toml / tui.toml, no environment variable, and no "don't ask again" choice in the dialog itself. Users who always want the same answer have to confirm every single goal.
Requested behavior
Any one of these would work:
- A "don't ask again" / "always keep YOLO" choice in the dialog that persists (for example under
~/.kimi-code) - A config key, e.g.
goal_start_prompt = falseor a[goal]section option inconfig.toml - Respect
default_permission_modeas the pre-approved answer and skip the prompt when it was explicitly configured
Why
Users who deliberately run in yolo mode have already accepted auto-approval of tool actions. The extra per-goal prompt adds a manual step to an otherwise autonomous workflow, especially when queuing goals regularly.
Environment
- kimi-code CLI version: 0.29.1
- OS: macOS
- Permission mode: yolo (set as default_permission_mode in config.toml)
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
Start at the /goal entry point and locate the TUI confirmation dialog described in the issue. Read how default_permission_mode is loaded from config.toml, then determine which requested persistence or configuration behavior fits the existing flow; done means a deliberate user choice or explicit setting prevents the prompt on later goal starts while other modes retain it.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- cli
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100