Repository navigation
ci: release packages - #4552
ci: release packages#4552
Conversation
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: b48d1a0c49
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
|
|
||
| ### Patch Changes | ||
|
|
||
| - [#4521](https://github.com/MoonshotAI/kimi-code/pull/4521) [`dbb65af`](https://github.com/MoonshotAI/kimi-code/commit/dbb65affd82525096c439e01d34e99d5dcf85318) Thanks [@7Sageer](https://github.com/7Sageer)! - Shorten built-in tool and subagent prompts. |
There was a problem hiding this comment.
When users install 2.4.1, the bundled dbb65af change removes the general instruction to read a target file before every Edit; a repo-wide search finds no replacement for ordinary editing sessions. Describing this only as “Shorten built-in tool and subagent prompts” hides that behavioral loss and provides neither an escape hatch nor an explanation of what now enforces the invariant for CLI users, so preserve the instruction or explicitly document and approve its removal before publishing.
AGENTS.md reference: AGENTS.md:L115-L121
Useful? React with 👍 / 👎.
b48d1a0 to
cd37f50
Compare
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: cd37f50604
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
|
|
||
| - [#4521](https://github.com/MoonshotAI/kimi-code/pull/4521) [`dbb65af`](https://github.com/MoonshotAI/kimi-code/commit/dbb65affd82525096c439e01d34e99d5dcf85318) Thanks [@7Sageer](https://github.com/7Sageer)! - Shorten built-in tool and subagent prompts. | ||
|
|
||
| - [#4521](https://github.com/MoonshotAI/kimi-code/pull/4521) [`dbb65af`](https://github.com/MoonshotAI/kimi-code/commit/dbb65affd82525096c439e01d34e99d5dcf85318) Thanks [@7Sageer](https://github.com/7Sageer)! - Shorten repeated goal and plan mode reminders. |
There was a problem hiding this comment.
For goal-mode and multi-turn plan-mode users, dbb65af changes GoalInjection to replace the full goal reminder after its first disclosure with goal-active-sparse-reminder.md, while the sparse plan reminders stop repeating the auto-permission fallback and multiple-approach guidance. Describing this only as “Shorten repeated goal and plan mode reminders” ships those prompt-behavior changes without naming what is no longer reinforced, what now enforces each safeguard, or an escape hatch or explicit maintainer sign-off; expand the release note and obtain approval before publishing.
AGENTS.md reference: AGENTS.md:L119-L121
Useful? React with 👍 / 👎.
cd37f50 to
6321d38
Compare
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 6321d381cc
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
|
|
||
| ### Patch Changes | ||
|
|
||
| - [#4553](https://github.com/MoonshotAI/kimi-code/pull/4553) [`9140ea0`](https://github.com/MoonshotAI/kimi-code/commit/9140ea0d980fc27c5c5ff983b2efe82191eb2b4a) Thanks [@RealKai42](https://github.com/RealKai42)! - A single file read by the agent now returns at most 100,000 characters or about 25,000 tokens, whichever comes first, instead of up to 500,000 characters; set `max_chars` under `[read]` in config.toml to raise both limits again. |
There was a problem hiding this comment.
For every agent that successfully uses Edit or Write, the bundled 9140ea0 change now injects guidance not to read the file back, and it also adds the same instruction to the Read tool description; this entry mentions only the new read-size cap, so it does not disclose that agent-behavior change or identify what now provides equivalent verification. Name this behavior and the affected population before publishing 2.4.1.
AGENTS.md reference: AGENTS.md:L119-L121
Useful? React with 👍 / 👎.
|
|
||
| - [#4521](https://github.com/MoonshotAI/kimi-code/pull/4521) [`dbb65af`](https://github.com/MoonshotAI/kimi-code/commit/dbb65affd82525096c439e01d34e99d5dcf85318) Thanks [@7Sageer](https://github.com/7Sageer)! - Shorten repeated goal and plan mode reminders. | ||
|
|
||
| - [#4555](https://github.com/MoonshotAI/kimi-code/pull/4555) [`480a9b0`](https://github.com/MoonshotAI/kimi-code/commit/480a9b0499b86d1421ce2c751cd8aaa798e2190a) Thanks [@kimi-agent-bot](https://github.com/kimi-agent-bot)! - web: Improved interactions and fixed known bugs. |
There was a problem hiding this comment.
For users deciding whether to install 2.4.1, “Improved interactions and fixed known bugs” identifies no observable behavior or fixed defect even though 480a9b0 replaces the shipped web bundle. Replace this with the concrete interaction changes and fixes so the published changeset states what actually changed.
AGENTS.md reference: AGENTS.md:L90-L91
Useful? React with 👍 / 👎.
This PR was opened by the Changesets release GitHub action. When you're ready to do a release, you can merge this and publish to npm yourself or setup this action to publish automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to release/2026-10-30, this PR will be updated.
Releases
@moonshot-ai/kimi-code@2.4.1
Patch Changes
#4553
9140ea0Thanks @RealKai42! - A single file read by the agent now returns at most 100,000 characters or about 25,000 tokens, whichever comes first, instead of up to 500,000 characters; setmax_charsunder[read]in config.toml to raise both limits again.#4521
dbb65afThanks @7Sageer! - Shorten built-in tool and subagent prompts.#4521
dbb65afThanks @7Sageer! - Shorten repeated goal and plan mode reminders.#4555
480a9b0Thanks @kimi-agent-bot! - web: Improved interactions and fixed known bugs.