docs: correct stale CLI flags/keybinding and document model.reasoningEffort - #6219
Merged
Merged
Conversation
…Effort - Remove nonexistent --all-files/-a and --show-memory-usage flags from the CLI arguments and headless option tables (no longer defined in the yargs parser in packages/cli/src/config/config.ts). - Add the commonly-needed --model/-m flag to the headless options table and fix the --approval-mode example to use the valid choice auto-edit (the parser rejects the underscore form auto_edit). - Drop the stale Meta+Enter alias from the external-editor shortcut; that chord is bound to NEWLINE, while OPEN_EXTERNAL_EDITOR binds only Ctrl+X. - Document the model.reasoningEffort setting (set via /effort), which is exposed in the settings dialog but was missing from the settings reference.
Collaborator
|
Thanks for the PR, @DragonnZhang! Template looks good ✓ On direction: straightforward docs fix — stale flags, wrong keybinding, and a missing setting entry. Keeping docs aligned with the parser is unambiguously valuable and squarely within scope. No concerns. On approach: the diff is minimal and focused — four surgical corrections across three files, each backed by a specific code reference. The scope discipline is excellent: the PR body explicitly calls out additional drift (voice settings, policy settings, missing keybindings) and defers them to a follow-up rather than bundling everything. That's the right call. Moving on to code review. 🔍 中文说明感谢贡献，@DragonnZhang！ 模板完整 ✓ 方向：纯粹的文档修正——过时的标志、错误的快捷键、缺失的设置项。保持文档与解析器一致毫无疑问是有价值的，完全在项目范围内。无顾虑。 方案：diff 极小且聚焦——三个文件中四处精准修正，每处都有具体代码引用。范围控制很好：PR body 明确提到了其他漂移（voice 设置、policy 设置、缺失快捷键）并留给后续 PR 处理，而非一股脑塞进来。这是正确的做法。 进入代码审查 🔍 — Qwen Code · qwen3.7-max |
Collaborator
Code ReviewIndependent baseline: for a "docs drifted from code" fix, I'd read the yargs parser in Verified each claim against the codebase:
No issues found. The diff is exactly what it claims to be — four precise corrections, nothing extra. TestingDocs-only change — no user-visible behavior to test via tmux. The "testing" here is the code verification above, which confirms every doc change matches the source of truth. N/A for Before/After. — Qwen Code · qwen3.7-max |
Collaborator
|
This is exactly what a docs-fix PR should look like. Four corrections, each one verifiable against the source, zero extras. The author clearly audited the code before writing — every claim in the PR body checks out when you go look at the parser, keybindings, and settings schema. The scope restraint deserves a callout: there's clearly more docs drift to fix (the PR body lists several other undocumented settings and keybindings), but the author chose to ship the high-confidence, unambiguous fixes now and defer the rest. That's the right instinct — small, correct PRs merge faster and cause fewer conflicts than big "while I'm here" bundles. All stages clean. Approving. ✅ 中文说明这就是文档修正 PR 该有的样子。四处修正，每处都可以对照源码验证，零多余内容。作者显然先审计了代码再动手——PR body 中的每个声明在解析器、快捷键和设置 schema 中都能确认。 值得特别提一下范围控制：显然还有更多文档漂移可以修（PR body 列出了其他几个未记录的设置和快捷键），但作者选择先把高确信度、无歧义的修正发出来，其余留待后续。这是正确的做法——小而正确的 PR 比大而全的更容易合并、更少冲突。 全部阶段通过，批准。✅ — Qwen Code · qwen3.7-max |
qwen-code-ci-bot
approved these changes
Jul 2, 2026
qwen-code-ci-bot
left a comment
qwen-code-ci-bot
left a comment
Collaborator
There was a problem hiding this comment.
LGTM, looks ready to ship. ✅
Contributor
Code Coverage Summary
CLI Package - Full Text ReportCore Package - Full Text ReportFor detailed HTML reports, please see the 'coverage-reports-22.x-ubuntu-latest' artifact from the main CI run. |
wenshao
approved these changes
Jul 3, 2026
| | Setting | Type | Description | Default | | ||
| | -------------------------------------------------- | ------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------- | | ||
| | `model.name` | string | The Qwen model to use for conversations. | `undefined` | | ||
| | `model.reasoningEffort` | enum | How hard reasoning-capable models think, applied across all providers. Set with the [`/effort`](../features/commands) command (`low`, `medium`, `high`, `xhigh`, `max`). Each provider maps and clamps this to what the active model supports (e.g. Gemini caps at `high`; Anthropic clamps tiers a model lacks). Leave unset to use the model/provider default. | `undefined` | |
Collaborator
There was a problem hiding this comment.
[Suggestion] The new model.reasoningEffort row has trailing whitespace that fails npx prettier --check. CI doesn't currently check Prettier on Markdown, but npm run format would auto-fix this after merge. Running npx prettier --write docs/users/configuration/settings.md before committing would keep things clean.
Suggested change
| | `model.reasoningEffort` | enum | How hard reasoning-capable models think, applied across all providers. Set with the [`/effort`](../features/commands) command (`low`, `medium`, `high`, `xhigh`, `max`). Each provider maps and clamps this to what the active model supports (e.g. Gemini caps at `high`; Anthropic clamps tiers a model lacks). Leave unset to use the model/provider default. | `undefined` | | |
| | `model.reasoningEffort` | enum | How hard reasoning-capable models think, applied across all providers. Set with the [`/effort`](../features/commands) command (`low`, `medium`, `high`, `xhigh`, `max`). Each provider maps and clamps this to what the active model supports (e.g. Gemini caps at `high`; Anthropic clamps tiers a model lacks). Leave unset to use the model/provider default. | `undefined` | |
— qwen3.7-max via Qwen Code /review
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What this PR does
Fixes documentation drift found by auditing
docs/against the current codebase. Four corrections, all backed by the code that is the source of truth: (1) removes the--all-files/-aand--show-memory-usageflags from the CLI-argument and headless option tables — neither is defined in the yargs parser anymore; (2) adds the commonly needed--model/-mflag to the headless options table and fixes the--approval-modeexample to use the valid choiceauto-edit(the parser rejects the underscore formauto_edit); (3) drops the staleMeta+Enteralias from the "open external editor" keyboard shortcut, since that chord is bound to newline while the external editor is bound only toCtrl+X; (4) documents themodel.reasoningEffortsetting, which is exposed in the settings dialog and set via/effortbut was missing from the settings reference.Why it's needed
These pages tell users which flags, shortcuts, and settings exist. Documenting flags that the CLI no longer accepts (
--all-files,--show-memory-usage) and an approval-mode value the parser rejects (auto_edit) leads users to commands that error out. TheMeta+Entermapping is wrong (it inserts a newline, not opens the editor). And a real, dialog-exposed setting (model.reasoningEffort) had no reference entry. Aligning the docs with the code removes these onboarding/usage papercuts.Reviewer Test Plan
How to verify
packages/cli/src/config/config.ts, the yargs parser defines--model/-mand restricts--approval-modeto['plan','default','auto-edit','auto','yolo']; there is noall-files/-aorshow-memory-usage/showMemoryUsageoption (confirmed absent repo-wide).qwen -p "hi" --all-filesandqwen --approval-mode auto_editboth fail against the current build.packages/cli/src/config/keyBindings.ts,Command.OPEN_EXTERNAL_EDITORbinds onlyCtrl+X({ key: 'x', ctrl: true }), whileMeta/Cmd+Enter({ key: 'return', command: true }) is bound toCommand.NEWLINE.packages/cli/src/config/settingsSchema.ts,model.reasoningEffortis an enum (low/medium/high/xhigh/max) withshowInDialog: true, set via the/effortcommand already documented indocs/users/features/commands.md.Evidence (Before & After)
N/A (docs-only change).
Tested on
Environment (optional)
N/A — docs-only; verified by reading the parser, keybinding, and settings schema sources.
Risk & Scope
general.voice.*,policy.*,agents.*,permissions.autoMode.*) and a couple of missing keyboard bindings (Alt+Ttoggle thinking,Ctrl+Fshell-input focus); those were left for a follow-up to keep this change focused on the highest-impact, unambiguous drift.Linked Issues
N/A
中文说明
本 PR 的作用
修复对照当前代码库审计
docs/时发现的文档漂移，共四处修正，均以作为事实来源的代码为依据：(1) 从命令行参数表和无头模式选项表中移除--all-files/-a与--show-memory-usage标志——两者在 yargs 解析器中都已不再定义；(2) 在无头模式选项表中加入常用的--model/-m标志，并将--approval-mode示例改为合法取值auto-edit（解析器会拒绝下划线形式的auto_edit）；(3) 从“打开外部编辑器”的快捷键中删除已失效的Meta+Enter别名，因为该组合键绑定的是换行，而外部编辑器仅绑定Ctrl+X；(4) 补充model.reasoningEffort设置项的文档，该设置会出现在设置对话框中并通过/effort命令设置，但此前在设置参考中缺失。为什么需要
这些页面告诉用户存在哪些标志、快捷键和设置。记录 CLI 已不再接受的标志（
--all-files、--show-memory-usage）以及解析器会拒绝的审批模式取值（auto_edit），会让用户执行报错的命令。Meta+Enter的映射是错误的（它插入换行，而非打开编辑器）。而一个真实存在、对话框中可见的设置（model.reasoningEffort）此前没有参考条目。使文档与代码保持一致可消除这些上手/使用中的小障碍。复核测试计划
如何验证
packages/cli/src/config/config.ts中，yargs 解析器定义了--model/-m，并将--approval-mode限定为['plan','default','auto-edit','auto','yolo']；不存在all-files/-a或show-memory-usage选项（已在全仓库确认缺失）。qwen -p "hi" --all-files与qwen --approval-mode auto_edit在当前构建下都会失败。packages/cli/src/config/keyBindings.ts中，Command.OPEN_EXTERNAL_EDITOR仅绑定Ctrl+X，而Meta/Cmd+Enter绑定到Command.NEWLINE。packages/cli/src/config/settingsSchema.ts中，model.reasoningEffort是枚举（low/medium/high/xhigh/max），showInDialog: true，通过已在docs/users/features/commands.md记录的/effort命令设置。证据（前后对比）
N/A（仅文档改动）。
测试平台
环境（可选）
N/A——仅文档改动；通过阅读解析器、快捷键与设置 schema 源码验证。
风险与范围
general.voice.*、policy.*、agents.*、permissions.autoMode.*）以及少量缺失的快捷键（Alt+T切换思考、Ctrl+Fshell 输入聚焦）；为使本次改动聚焦于影响最大、最明确的漂移，这些留待后续处理。关联 Issue
N/A
Generated by Claude Code