@qwen-code/qwen-code 0.18.3 → 0.18.4-preview.0

package/bundled/loop/SKILL.md

@@ -1,14 +1,15 @@
 ---
 name: loop
-description: Create a recurring loop that runs a prompt on a schedule. Usage - /loop 5m check the build, /loop check the PR every 30m, /loop run tests (defaults to 10m). /loop list to show jobs, /loop clear to cancel all.
-argument-hint: '[interval] <prompt> | list | clear'
+description: Create a loop that runs a prompt now and follows up either on a fixed schedule or through self-paced wakeups. Usage - /loop check the build, /loop 5m check the build, /loop check the PR every 30m. /loop list to show jobs, /loop clear to cancel all.
+argument-hint: '[interval] [prompt] | list | clear'
 allowedTools:
   - cron_create
   - cron_list
   - cron_delete
+  - loop_wakeup
 ---
 
-# /loop — schedule a recurring prompt
+# /loop — run a prompt repeatedly
 
 ## Subcommands
 
@@ -17,47 +18,73 @@ If the input (after stripping the `/loop` prefix) is exactly one of these keywor
 - **`list`** — call CronList and display the results. Done.
 - **`clear`** — call CronList, then call CronDelete for every job returned. Confirm how many were cancelled. Done.
 
-Otherwise, parse the input below into `[interval] <prompt…>` and schedule it with CronCreate.
+## Parsing
 
-## Parsing (in priority order)
+Parse the input after removing the `/loop` prefix:
 
-1. **Leading token**: if the first whitespace-delimited token matches `^\d+[smhd]$` (e.g. `5m`, `2h`), that's the interval; the rest is the prompt.
-2. **Trailing "every" clause**: otherwise, if the input ends with `every <N><unit>` or `every <N> <unit-word>` (e.g. `every 20m`, `every 5 minutes`, `every 2 hours`), extract that as the interval and strip it from the prompt. Only match when what follows "every" is a time expression — `check every PR` has no interval.
-3. **Default**: otherwise, interval is `10m` and the entire input is the prompt.
+1. **Empty input**: show usage `/loop [interval] [prompt]` and stop. Do not call CronCreate or LoopWakeup in this slice.
+2. **Leading interval token**: if the first whitespace-delimited token matches `^\d+[smhd]$` (e.g. `5m`, `2h`), this is the fixed-interval recurring path. The rest is the prompt.
+3. **Trailing "every" clause**: otherwise, if the input ends with `every <N><unit>` or `every <N> <unit-word>` (e.g. `every 20m`, `every 5 minutes`, `every 2 hours`), this is the fixed-interval recurring path. Extract that interval and strip it from the prompt. Only match when what follows "every" is a time expression — `check every PR` has no interval.
+4. **Prompt-only input**: otherwise, the entire input is the prompt and this is the prompt-only self-paced path.
 
-If the resulting prompt is empty, show usage `/loop [interval] <prompt>` and stop — do not call CronCreate.
+If the resulting prompt is empty, show usage `/loop [interval] [prompt]` and stop.
 
 Examples:
 
-- `5m /babysit-prs` → interval `5m`, prompt `/babysit-prs` (rule 1)
-- `check the deploy every 20m` → interval `20m`, prompt `check the deploy` (rule 2)
-- `run tests every 5 minutes` → interval `5m`, prompt `run tests` (rule 2)
-- `check the deploy` → interval `10m`, prompt `check the deploy` (rule 3)
-- `check every PR` → interval `10m`, prompt `check every PR` (rule 3 — "every" not followed by time)
+- `5m /babysit-prs` → fixed interval `5m`, prompt `/babysit-prs` (leading interval token)
+- `check the deploy every 20m` → fixed interval `20m`, prompt `check the deploy` (trailing "every" clause)
+- `run tests every 5 minutes` → fixed interval `5m`, prompt `run tests` (trailing "every" clause)
+- `check every PR` → prompt-only self-paced path, prompt `check every PR` ("every" is not followed by a time expression)
+- `check the deploy` → prompt-only self-paced path, prompt `check the deploy`
 - `5m` → empty prompt → show usage
 
-## Interval → cron
+## Prompt-only self-paced path
+
+Use this path only when the user supplied a prompt and no interval.
+
+1. Do not call CronCreate for this path.
+2. Run the parsed prompt immediately now.
+   - If it is a slash command, invoke it via the Skill tool.
+   - Otherwise, act on it directly.
+3. Before ending the turn, decide whether another check is useful.
+   - Call LoopWakeup only if continued follow-up is useful.
+   - Do not call LoopWakeup if the task is complete.
+   - Do not call LoopWakeup if the task is blocked on user input or external state that cannot be checked later.
+   - Do not call LoopWakeup just to keep polling when no useful next check exists.
+4. When scheduling a continuation, call LoopWakeup with:
+   - `delaySeconds`: the next useful delay in seconds. The runtime clamps to 60–3600 (1–60 min); follow the tool's own guidance on picking a value (it accounts for the prompt-cache window).
+   - `prompt`: `/loop ${original prompt}`
+   - `reason`: a short reason for the chosen delay.
+5. Briefly tell the user what was done now. If a wakeup was scheduled, include when the next check is expected. If no wakeup was scheduled, say the loop is complete or not continuing.
+
+## Fixed-interval recurring path
+
+Use this path only for inputs with a leading interval token or a trailing "every" clause.
+
+### Interval to cron
 
 Supported suffixes: `s` (seconds, rounded up to nearest minute, min 1), `m` (minutes), `h` (hours), `d` (days). Convert:
 
-| Interval pattern  | Cron expression        | Notes                                     |
-| ----------------- | ---------------------- | ----------------------------------------- |
-| `Nm` where N ≤ 59 | `*/N * * * *`          | every N minutes                           |
-| `Nm` where N ≥ 60 | `0 */H * * *`          | round to hours (H = N/60, must divide 24) |
-| `Nh` where N ≤ 23 | `0 */N * * *`          | every N hours                             |
-| `Nd`              | `0 0 */N * *`          | every N days at midnight local            |
-| `Ns`              | treat as `ceil(N/60)m` | cron minimum granularity is 1 minute      |
+| Interval pattern   | Cron expression        | Notes                                     |
+| ------------------ | ---------------------- | ----------------------------------------- |
+| `Nm` where N <= 59 | `*/N * * * *`          | every N minutes                           |
+| `Nm` where N >= 60 | `0 */H * * *`          | round to hours (H = N/60, must divide 24) |
+| `Nh` where N <= 23 | `0 */N * * *`          | every N hours                             |
+| `Nd`               | `0 0 */N * *`          | every N days at midnight local            |
+| `Ns`               | treat as `ceil(N/60)m` | cron minimum granularity is 1 minute      |
 
-**If the interval doesn't cleanly divide its unit** (e.g. `7m` → `*/7 * * * *` gives uneven gaps at :56→:00; `90m` → 1.5h which cron can't express), pick the nearest clean interval and tell the user what you rounded to before scheduling.
+If the interval does not cleanly divide its unit (for example `7m` gives uneven gaps at `:56` to `:00`, or `90m` is 1.5 hours which cron cannot express), pick the nearest clean interval and tell the user what you rounded to before scheduling.
 
-## Action
+### Action
 
 1. Call CronCreate with:
    - `cron`: the expression from the table above
    - `prompt`: the parsed prompt from above, verbatim (slash commands are passed through unchanged)
    - `recurring`: `true`
    - `durable`: `true` if the user's language implies persistence ("keep doing this", "set this up permanently", "every day even after restart"). Otherwise omit (defaults to session-only).
-2. Briefly confirm: what's scheduled, the cron expression, the human-readable cadence, that recurring tasks auto-expire after 7 days, and that they can cancel sooner with CronDelete (include the job ID).
-3. **Then immediately execute the parsed prompt now** — don't wait for the first cron fire. If it's a slash command, invoke it via the Skill tool; otherwise act on it directly.
+2. Briefly confirm: what is scheduled, the cron expression, the human-readable cadence, that recurring tasks auto-expire after 7 days, and that they can cancel sooner with CronDelete (include the job ID).
+3. Then immediately execute the parsed prompt now. Do not wait for the first cron fire.
+   - If it is a slash command, invoke it via the Skill tool.
+   - Otherwise, act on it directly.
 
 ## Input

package/bundled/loop/SKILL.test.ts

@@ -0,0 +1,67 @@
+/**
+ * @license
+ * Copyright 2026 Qwen
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import * as fs from 'node:fs';
+import * as path from 'node:path';
+import { fileURLToPath } from 'node:url';
+import { describe, expect, it } from 'vitest';
+import { parseSkillContent } from '../../skill-load.js';
+
+function loadLoopSkill() {
+  const skillPath = path.join(
+    path.dirname(fileURLToPath(import.meta.url)),
+    'SKILL.md',
+  );
+  const content = fs.readFileSync(skillPath, 'utf8');
+  const config = parseSkillContent(content, skillPath);
+  return { config, body: config.body };
+}
+
+describe('bundled loop skill', () => {
+  it('allows both cron tools and the loop wakeup primitive', () => {
+    const { config } = loadLoopSkill();
+
+    expect(config.allowedTools).toEqual([
+      'cron_create',
+      'cron_list',
+      'cron_delete',
+      'loop_wakeup',
+    ]);
+  });
+
+  it('documents prompt-only input as self-paced and non-recurring', () => {
+    const { body } = loadLoopSkill();
+
+    expect(body).toContain('## Prompt-only self-paced path');
+    expect(body).toContain('Do not call CronCreate for this path.');
+    expect(body).toContain('Run the parsed prompt immediately now.');
+    expect(body).toContain(
+      'Call LoopWakeup only if continued follow-up is useful.',
+    );
+    expect(body).toContain('`prompt`: `/loop ${original prompt}`');
+    // Drives the second-resolution wakeup engine — seconds, not minutes.
+    expect(body).toContain('`delaySeconds`');
+    expect(body).not.toContain('delayMinutes');
+  });
+
+  it('keeps fixed-interval inputs on the recurring cron path', () => {
+    const { body } = loadLoopSkill();
+
+    expect(body).toContain('## Fixed-interval recurring path');
+    expect(body).toContain('Leading interval token');
+    expect(body).toContain('Trailing "every" clause');
+    expect(body).toContain('Call CronCreate with:');
+    expect(body).toContain('recurring`: `true`');
+  });
+
+  it('keeps list and clear as management subcommands', () => {
+    const { body } = loadLoopSkill();
+
+    expect(body).toContain('**`list`** — call CronList');
+    expect(body).toContain('**`clear`** — call CronList');
+    expect(body).toContain('call CronDelete for every job returned');
+  });
+});

package/bundled/qc-helper/docs/configuration/settings.md

@@ -111,13 +111,15 @@ Settings are organized into categories. Most settings should be placed within th
 | `ui.showLineNumbers`                    | boolean          | Show line numbers in code blocks in the CLI output.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                | `true`        |
 | `ui.renderMode`                         | string           | Default Markdown display mode. Use `"render"` for rich visual previews or `"raw"` to show source-oriented Markdown by default. Toggle during a session with `Alt/Option+M`; on macOS the terminal must send Option as Meta. See [Markdown Rendering](../features/markdown-rendering).                                                                                                                                                                                                                                                              | `"render"`    |
 | `ui.showCitations`                      | boolean          | Show citations for generated text in the chat.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     | `false`       |
+| `ui.history.collapseOnResume`           | boolean          | Whether to collapse history by default when resuming a session. Can be toggled via `/history collapse-on-resume` and `/history expand-on-resume`.                                                                                                                                                                                                                                                                                                                                                                                                  | `false`       |
 | `ui.compactMode`                        | boolean          | Hide tool output and thinking for a cleaner view. Toggle with `Ctrl+O` during a session or via the Settings dialog. Tool approval prompts are never hidden, even in compact mode. The setting persists across sessions.                                                                                                                                                                                                                                                                                                                            | `false`       |
 | `ui.shellOutputMaxLines`                | number           | Max number of shell output lines shown inline. Set to `0` to disable the cap and show full output. Hidden lines are surfaced via the `+N lines` indicator. Errors, `!`-prefix user-initiated commands, confirming tools, and focused embedded shells always show full output.                                                                                                                                                                                                                                                                      | `5`           |
 | `ui.enableWelcomeBack`                  | boolean          | Show welcome back dialog when returning to a project with conversation history. When enabled, Qwen Code will automatically detect if you're returning to a project with a previously generated project summary (`.qwen/PROJECT_SUMMARY.md`) and show a dialog allowing you to continue your previous conversation or start fresh. If you choose **Start new chat session**, that choice is remembered for the current project until the project summary changes. This feature integrates with the `/summary` command and quit confirmation dialog. | `true`        |
 | `ui.accessibility.enableLoadingPhrases` | boolean          | Enable loading phrases (disable for accessibility).                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                | `true`        |
 | `ui.accessibility.screenReader`         | boolean          | Enables screen reader mode, which adjusts the TUI for better compatibility with screen readers.                                                                                                                                                                                                                                                                                                                                                                                                                                                    | `false`       |
 | `ui.customWittyPhrases`                 | array of strings | A list of custom phrases to display during loading states. When provided, the CLI will cycle through these phrases instead of the default ones.                                                                                                                                                                                                                                                                                                                                                                                                    | `[]`          |
-| `ui.enableFollowupSuggestions`          | boolean          | Enable [followup suggestions](../features/followup-suggestions) that predict what you want to type next after the model responds. Suggestions appear as ghost text and can be accepted with Tab, Enter, or Right Arrow.                                                                                                                                                                                                                                                                                                                            | `false`       |
+| `ui.showResponseTokensPerSecond`        | boolean          | Show a live tokens/sec estimate next to the response token counter while the model is streaming. This is a generation-speed hint, not an ETA or completion percentage. Takes effect in the next session.                                                                                                                                                                                                                                                                                                                                           | `false`       |
+| `ui.enableFollowupSuggestions`          | boolean          | Enable [followup suggestions](../features/followup-suggestions) that predict what you want to type next after the model responds. Suggestions appear as placeholder text and are accepted with Tab, Enter, or Right Arrow (which fill the input — they do not auto-submit). On by default; set to `false` to opt out.                                                                                                                                                                                                                              | `true`        |
 | `ui.enableCacheSharing`                 | boolean          | Use cache-aware forked queries for suggestion generation. Reduces cost on providers that support prefix caching (experimental).                                                                                                                                                                                                                                                                                                                                                                                                                    | `true`        |
 | `ui.enableSpeculation`                  | boolean          | Speculatively execute accepted suggestions before submission. Results appear instantly when you accept (experimental).                                                                                                                                                                                                                                                                                                                                                                                                                             | `false`       |
 | `experimental.emitToolUseSummaries`     | boolean          | Generate short LLM-based labels summarizing each tool-call batch. See [Tool-Use Summaries](../features/tool-use-summaries). Requires `fastModel` to be configured; silently skipped otherwise. Can be overridden per-session with `QWEN_CODE_EMIT_TOOL_USE_SUMMARIES=0` or `=1`.                                                                                                                                                                                                                                                                   | `true`        |
@@ -137,23 +139,23 @@ Settings are organized into categories. Most settings should be placed within th
 
 #### model
 
-| Setting                                            | Type    | Description                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                            | Default     |
-| -------------------------------------------------- | ------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------- |
-| `model.name`                                       | string  | The Qwen model to use for conversations.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               | `undefined` |
-| `model.maxSessionTurns`                            | number  | Maximum number of user/model/tool turns to keep in a session. -1 means unlimited.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      | `-1`        |
-| `model.maxWallTimeSeconds`                         | number  | Wall-clock budget for headless / unattended runs, in seconds. `-1` means unlimited. Overridable per-invocation via `--max-wall-time`, which requires a positive duration (`90`, `30s`, `5m`, `1h`, `1.5h`); the minimum is 1 second — sub-second values (`500ms`, `0.5`) are rejected as typos. Omit the flag to fall back to this setting. Aborts with exit code 55 when exceeded.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    | `-1`        |
-| `model.maxToolCalls`                               | number  | Cumulative tool-call budget for a run (counts every executed tool, success or failure; `structured_output` under `--json-schema` is exempt). `-1` means unlimited; `0` means "no tool calls allowed". Capped at 1,000,000 to catch typos. Overridable via `--max-tool-calls`. Aborts with exit code 55 when exceeded.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                  | `-1`        |
-| `model.generationConfig`                           | object  | Advanced overrides passed to the underlying content generator. Supports request controls such as `timeout`, `maxRetries`, `enableCacheControl`, `splitToolMedia` (default `true`; splits tool-returned media — including images read by the built-in read_file — into a follow-up user message instead of the spec-violating `role: "tool"` message, so strict OpenAI-compatible servers like doubao / new-api / LM Studio can see it; set `false` to restore the legacy embed-in-tool behavior), `contextWindowSize` (override model's context window size), `modalities` (override auto-detected input modalities), `customHeaders` (custom HTTP headers for API requests), and `extra_body` (additional body parameters for OpenAI-compatible API requests only), along with fine-tuning knobs under `samplingParams` (for example `temperature`, `top_p`, `max_tokens`). Leave unset to rely on provider defaults. | `undefined` |
-| `model.chatCompression.contextPercentageThreshold` | number  | **REMOVED.** Auto-compaction now uses a three-tier threshold ladder (warn / auto / hard) computed internally from the model's context window via the `computeThresholds()` function — no longer user-configurable. Setting this field in `settings.json` is silently ignored (no startup warning). There is currently no replacement for "disable compression entirely" — reactive overflow recovery remains the safety net at the API layer if compression itself fails. (See PR #4345 / `docs/design/auto-compaction-threshold-redesign.md` for the redesign rationale.)                                                                                                                                                                                                                                                                                                                                             | `N/A`       |
-| `model.chatCompression.maxRecentFilesToRetain`     | number  | Number of most-recently-touched files whose current content is restored (embedded if small, otherwise referenced by path) into history after auto-compaction. `0` restores none. Env override: `QWEN_COMPACT_MAX_RECENT_FILES`.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                        | `5`         |
-| `model.chatCompression.maxRecentImagesToRetain`    | number  | Number of most-recent images (tool screenshots / user pastes) restored into history after auto-compaction. `0` restores none. Env override: `QWEN_COMPACT_MAX_RECENT_IMAGES`.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                          | `3`         |
-| `model.chatCompression.enableScreenshotTrigger`    | boolean | When `true`, auto-compaction also fires once the number of tool-returned images accumulated in history reaches `screenshotTriggerThreshold`, independent of token usage — aimed at computer-use sessions where frequent screenshots dilute model attention. Counts only images returned inside tool results, not user-pasted images. Env override: `QWEN_COMPACT_SCREENSHOT_TRIGGER` (`1`/`true`/`0`/`false`).                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         | `true`      |
-| `model.chatCompression.screenshotTriggerThreshold` | number  | Tool-returned image count at or above which the screenshot trigger fires (only when `enableScreenshotTrigger`). Compaction resets the count — surviving images are re-embedded as top-level parts, which the trigger doesn't count — so it won't immediately re-fire. Env override: `QWEN_COMPACT_SCREENSHOT_THRESHOLD`.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               | `50`        |
-| `model.skipNextSpeakerCheck`                       | boolean | Skip the next speaker check.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           | `true`      |
-| `model.skipLoopDetection`                          | boolean | Disables streaming loop detection checks. Defaults to `true` (loop detection is skipped) to avoid false positives interrupting legitimate workflows. Set to `false` to re-enable streaming loop detection — useful as a guardrail in headless / non-interactive runs where stuck repetition can otherwise waste budget.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                | `true`      |
-| `model.skipStartupContext`                         | boolean | Skips sending the startup workspace context (environment summary and acknowledgement) at the beginning of each session. Enable this if you prefer to provide context manually or want to save tokens on startup.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       | `false`     |
-| `model.enableOpenAILogging`                        | boolean | Enables logging of OpenAI API calls for debugging and analysis. When enabled, API requests and responses are logged to JSON files.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     | `false`     |
-| `model.openAILoggingDir`                           | string  | Custom directory path for OpenAI API logs. If not specified, defaults to `logs/openai` in the current working directory. Supports absolute paths, relative paths (resolved from current working directory), and `~` expansion (home directory).                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                        | `undefined` |
+| Setting                                            | Type    | Description                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     | Default     |
+| -------------------------------------------------- | ------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------- |
+| `model.name`                                       | string  | The Qwen model to use for conversations.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                        | `undefined` |
+| `model.maxSessionTurns`                            | number  | Maximum number of user/model/tool turns to keep in a session. -1 means unlimited.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               | `-1`        |
+| `model.maxWallTimeSeconds`                         | number  | Wall-clock budget for headless / unattended runs, in seconds. `-1` means unlimited. Overridable per-invocation via `--max-wall-time`, which requires a positive duration (`90`, `30s`, `5m`, `1h`, `1.5h`); the minimum is 1 second — sub-second values (`500ms`, `0.5`) are rejected as typos. Omit the flag to fall back to this setting. Aborts with exit code 55 when exceeded.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             | `-1`        |
+| `model.maxToolCalls`                               | number  | Cumulative tool-call budget for a run (counts every executed tool, success or failure; `structured_output` under `--json-schema` is exempt). `-1` means unlimited; `0` means "no tool calls allowed". Capped at 1,000,000 to catch typos. Overridable via `--max-tool-calls`. Aborts with exit code 55 when exceeded.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           | `-1`        |
+| `model.generationConfig`                           | object  | Advanced overrides passed to the underlying content generator. Supports request controls such as `timeout`, `maxRetries`, `enableCacheControl`, `splitToolMedia` (default `true`; splits tool-returned media — including images read by the built-in read_file — into a follow-up user message instead of the spec-violating `role: "tool"` message, so strict OpenAI-compatible servers like doubao / new-api / LM Studio can see it; set `false` to restore the legacy embed-in-tool behavior), `toolResultContentFormat` (default `"parts"`; set `"string"` only for legacy OpenAI-compatible runtimes whose tool templates ignore text content parts), `contextWindowSize` (override model's context window size), `modalities` (override auto-detected input modalities), `customHeaders` (custom HTTP headers for API requests), and `extra_body` (additional body parameters for OpenAI-compatible API requests only), along with fine-tuning knobs under `samplingParams` (for example `temperature`, `top_p`, `max_tokens`). Leave unset to rely on provider defaults. | `undefined` |
+| `model.chatCompression.contextPercentageThreshold` | number  | **REMOVED.** Auto-compaction now uses a three-tier threshold ladder (warn / auto / hard) computed internally from the model's context window via the `computeThresholds()` function — no longer user-configurable. Setting this field in `settings.json` is silently ignored (no startup warning). There is currently no replacement for "disable compression entirely" — reactive overflow recovery remains the safety net at the API layer if compression itself fails. (See PR #4345 / `docs/design/auto-compaction-threshold-redesign.md` for the redesign rationale.)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      | `N/A`       |
+| `model.chatCompression.maxRecentFilesToRetain`     | number  | Number of most-recently-touched files whose current content is restored (embedded if small, otherwise referenced by path) into history after auto-compaction. `0` restores none. Env override: `QWEN_COMPACT_MAX_RECENT_FILES`.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 | `5`         |
+| `model.chatCompression.maxRecentImagesToRetain`    | number  | Number of most-recent images (tool screenshots / user pastes) restored into history after auto-compaction. `0` restores none. Env override: `QWEN_COMPACT_MAX_RECENT_IMAGES`.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   | `3`         |
+| `model.chatCompression.enableScreenshotTrigger`    | boolean | When `true`, auto-compaction also fires once the number of tool-returned images accumulated in history reaches `screenshotTriggerThreshold`, independent of token usage — aimed at computer-use sessions where frequent screenshots dilute model attention. Counts only images returned inside tool results, not user-pasted images. Env override: `QWEN_COMPACT_SCREENSHOT_TRIGGER` (`1`/`true`/`0`/`false`).                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                  | `true`      |
+| `model.chatCompression.screenshotTriggerThreshold` | number  | Tool-returned image count at or above which the screenshot trigger fires (only when `enableScreenshotTrigger`). Compaction resets the count — surviving images are re-embedded as top-level parts, which the trigger doesn't count — so it won't immediately re-fire. Env override: `QWEN_COMPACT_SCREENSHOT_THRESHOLD`.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                        | `50`        |
+| `model.skipNextSpeakerCheck`                       | boolean | Skip the next speaker check.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    | `true`      |
+| `model.skipLoopDetection`                          | boolean | Disables streaming loop detection checks. Defaults to `true` (loop detection is skipped) to avoid false positives interrupting legitimate workflows. Set to `false` to re-enable streaming loop detection — useful as a guardrail in headless / non-interactive runs where stuck repetition can otherwise waste budget.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         | `true`      |
+| `model.skipStartupContext`                         | boolean | Skips sending the startup workspace context (environment summary and acknowledgement) at the beginning of each session. Enable this if you prefer to provide context manually or want to save tokens on startup.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                | `false`     |
+| `model.enableOpenAILogging`                        | boolean | Enables logging of OpenAI API calls for debugging and analysis. When enabled, API requests and responses are logged to JSON files.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                              | `false`     |
+| `model.openAILoggingDir`                           | string  | Custom directory path for OpenAI API logs. If not specified, defaults to `logs/openai` in the current working directory. Supports absolute paths, relative paths (resolved from current working directory), and `~` expansion (home directory).                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 | `undefined` |
 
 **Example model.generationConfig:**
 
@@ -167,6 +169,7 @@ Settings are organized into categories. Most settings should be placed within th
         "image": true
       },
       "enableCacheControl": true,
+      "toolResultContentFormat": "parts",
       "customHeaders": {
         "X-Client-Request-ID": "req-123"
       },
@@ -195,6 +198,10 @@ This is transparent to users — you may briefly see a retry indicator if escala
 
 To override this behavior, either set `samplingParams.max_tokens` in your settings or use the `QWEN_CODE_MAX_OUTPUT_TOKENS` environment variable.
 
+**toolResultContentFormat:**
+
+Controls how text-only tool results are serialized in OpenAI-compatible requests. The default `"parts"` keeps the standard content-part array shape. Set `"string"` only for legacy OpenAI-compatible runtimes whose tool templates ignore text content parts, such as older GLM-5.1 vLLM/SGLang templates. Tool-returned media is still controlled by `splitToolMedia`.
+
 **contextWindowSize:**
 
 Overrides the default context window size for the selected model. Qwen Code determines the context window using built-in defaults based on model name matching, with a constant fallback value. Use this setting when a provider's effective context limit differs from Qwen Code's default. This value defines the model's assumed maximum context capacity, not a per-request token limit.

package/bundled/qc-helper/docs/extension/introduction.md

@@ -18,6 +18,16 @@ You can manage extensions at runtime within the interactive CLI using `/extensio
 | `/extensions install <source>`        | Install an extension from a git URL, local path, npm package, or marketplace |
 | `/extensions explore [source]`        | Open extensions source page(Gemini or ClaudeCode) in your browser            |
 
+#### The interactive extension manager
+
+Running `/extensions` (or `/extensions manage`) opens an interactive manager with three tabs. Press `Tab` or the `←`/`→` arrows to switch between them.
+
+- **Discover** — browse plugins from your configured marketplace sources. Type to search, `Enter` to view a plugin's details, and install it (you'll be asked to choose an install scope). Press `Ctrl+R` to re-fetch the listings, and `Esc` to go back.
+- **Installed** — your installed extensions, grouped by scope (**User level**, **Project level**, and favorites). Use `↑`/`↓` to navigate, `Space` to enable/disable an extension, `f` to favorite it, and `Enter` to open its details. MCP servers bundled by an extension appear nested under their parent extension with live connection status; you can enable or disable each server individually from there.
+- **Sources** — manage the marketplace sources that feed the Discover tab. Use `↑`/`↓` to navigate, `Enter` to select a source, and `d` to remove one. These are the same sources managed by the `qwen extensions sources` CLI commands described below.
+
+Changes made here hot-reload immediately, without restarting Qwen Code.
+
 ### CLI Extension Management
 
 You can also manage extensions using `qwen extensions` CLI commands. Note that changes made via CLI commands will be reflected in active CLI sessions on restart.
@@ -133,6 +143,34 @@ qwen extensions install /path/to/your/extension
 
 Note that we create a copy of the installed extension, so you will need to run `qwen extensions update` to pull in changes from both locally-defined extensions and those on GitHub.
 
+#### Choosing an install scope
+
+By default, an installed extension is enabled globally (user scope). Pass `--scope project` to enable it only for the current workspace:
+
+```bash
+qwen extensions install <source> --scope project
+```
+
+`--scope workspace` is accepted as an alias of `--scope project`. This matches the scope choice offered when installing from the `/extensions manage` Discover tab.
+
+### Managing marketplace sources
+
+Marketplace sources (Claude plugin marketplaces) power the Discover tab in `/extensions manage`. You can manage them from the CLI as well:
+
+```bash
+# Add a marketplace (owner/repo, git URL, https URL to marketplace.json, or local path)
+qwen extensions sources add <source>
+
+# List configured marketplaces
+qwen extensions sources list
+
+# Re-fetch a marketplace's plugin listing
+qwen extensions sources update <name>
+
+# Remove a marketplace
+qwen extensions sources remove <name>
+```
+
 ### Uninstalling an extension
 
 To uninstall, run `qwen extensions uninstall extension-name`, so, in the case of the install example:

package/bundled/qc-helper/docs/features/channels/_meta.ts

@@ -4,5 +4,6 @@ export default {
   weixin: 'WeChat',
   dingtalk: 'DingTalk',
   feishu: 'Feishu',
+  qqbot: 'QQ Bot',
   plugins: 'Plugins',
 };

package/bundled/qc-helper/docs/features/channels/overview.md

@@ -1,6 +1,6 @@
 # Channels
 
-Channels let you interact with a Qwen Code agent from messaging platforms like Telegram, WeChat, or DingTalk, instead of the terminal. You send messages from your phone or desktop chat app, and the agent responds just like it would in the CLI.
+Channels let you interact with a Qwen Code agent from messaging platforms like Telegram, WeChat, QQ, or DingTalk, instead of the terminal. You send messages from your phone or desktop chat app, and the agent responds just like it would in the CLI.
 
 ## How It Works
 
@@ -15,7 +15,7 @@ All channels share one agent process with isolated sessions per user. Each chann
 
 ## Quick Start
 
-1. Set up a bot on your messaging platform (see channel-specific guides: [Telegram](./telegram), [WeChat](./weixin), [DingTalk](./dingtalk))
+1. Set up a bot on your messaging platform (see channel-specific guides: [Telegram](./telegram), [WeChat](./weixin), [QQ Bot](./qqbot), [DingTalk](./dingtalk))
 2. Add the channel configuration to `~/.qwen/settings.json`
 3. Run `qwen channel start` to start all channels, or `qwen channel start <name>` for a single channel
 
@@ -49,7 +49,7 @@ Channels are configured under the `channels` key in `settings.json`. Each channe
 
 | Option                   | Required | Description                                                                                                                                    |
 | ------------------------ | -------- | ---------------------------------------------------------------------------------------------------------------------------------------------- |
-| `type`                   | Yes      | Channel type: `telegram`, `weixin`, `dingtalk`, or a custom type from an extension (see [Plugins](./plugins))                                  |
+| `type`                   | Yes      | Channel type: `telegram`, `weixin`, `qq`, `dingtalk`, `feishu`, or a custom type from an extension (see [Plugins](./plugins))                  |
 | `token`                  | Telegram | Bot token. Supports `$ENV_VAR` syntax to read from environment variables. Not needed for WeChat or DingTalk                                    |
 | `clientId`               | DingTalk | DingTalk AppKey. Supports `$ENV_VAR` syntax                                                                                                    |
 | `clientSecret`           | DingTalk | DingTalk AppSecret. Supports `$ENV_VAR` syntax                                                                                                 |
@@ -292,7 +292,7 @@ Channels support slash commands. These are handled locally (no agent round-trip)
 
 All other slash commands (e.g., `/compress`, `/summary`) are forwarded to the agent.
 
-These commands work on all channel types (Telegram, WeChat, DingTalk).
+These commands work on all channel types (Telegram, WeChat, QQ, DingTalk).
 
 ## Running
 

package/bundled/qc-helper/docs/features/channels/qqbot.md

@@ -0,0 +1,179 @@
+# QQ Bot (QQ机器人)
+
+This guide covers setting up a Qwen Code channel on QQ via the official QQ Bot Open Platform API.
+
+## Prerequisites
+
+- A QQ account (mobile app for scanning the QR code)
+
+## Setup
+
+### QR Code Login
+
+Start the channel — the first time it will show a QR code. Scan it with your QQ app to activate. No developer account or manual registration needed. Credentials are saved and reused automatically.
+
+```json
+{
+  "channels": {
+    "my-qq": {
+      "type": "qq"
+    }
+  }
+}
+```
+
+```bash
+qwen channel start my-qq
+# Scan the QR code in the terminal with your QQ app
+```
+
+### Manual Configuration (Developer Portal)
+
+You can also use credentials from the [QQ Bot Open Platform](https://q.qq.com/) developer portal if you already have an app registered there:
+
+```json
+{
+  "channels": {
+    "my-qq": {
+      "type": "qq",
+      "appID": "YOUR_APP_ID",
+      "appSecret": "$QQ_APP_SECRET"
+    }
+  }
+}
+```
+
+Set the secret as an environment variable:
+
+```bash
+export QQ_APP_SECRET=<your-app-secret>
+```
+
+## Configuration
+
+```json
+{
+  "channels": {
+    "my-qq": {
+      "type": "qq",
+      "appID": "YOUR_APP_ID",
+      "appSecret": "$QQ_APP_SECRET",
+      "sandbox": false,
+      "senderPolicy": "open",
+      "sessionScope": "user",
+      "cwd": "/path/to/your/project",
+      "instructions": "你是一个通过 QQ Bot 对话的 AI 助手。回复控制在 2000 字符以内。",
+      "blockStreaming": "on",
+      "groupPolicy": "disabled",
+      "groups": {
+        "*": { "requireMention": true }
+      }
+    }
+  }
+}
+```
+
+### QQ-Specific Options
+
+| Option      | Default | Description                                                                       |
+| ----------- | ------- | --------------------------------------------------------------------------------- |
+| `appID`     | —       | QQ Bot AppID from developer portal. If omitted, QR code login is used.            |
+| `appSecret` | —       | QQ Bot AppSecret. Supports `$ENV_VAR` syntax. If omitted, QR code login is used.  |
+| `sandbox`   | `false` | Set to `true` to use the QQ sandbox API environment (`sandbox.api.sgroup.qq.com`) |
+
+All standard channel options (see [Channel Overview](./overview#options)) are also supported:
+`senderPolicy`, `allowedUsers`, `sessionScope`, `cwd`, `instructions`, `groupPolicy`, `groups`, `dispatchMode`, `blockStreaming`, `blockStreamingChunk`, `blockStreamingCoalesce`.
+
+## Running
+
+```bash
+# Start only the QQ channel
+qwen channel start my-qq
+
+# Or start all configured channels together
+qwen channel start
+```
+
+Open QQ and send a message to your bot. You should see the response arrive in your chat.
+
+## Group Chats
+
+To use the bot in QQ groups:
+
+1. Set `groupPolicy` to `"allowlist"` or `"open"` in your channel config
+2. Add the bot to a QQ group via the QQ Bot Open Platform dashboard or by having a group admin invite it
+3. Group members must **@mention** the bot to trigger a response
+
+QQ Bot API V2 only delivers group messages that @mention the bot — the bot does not see all group messages. By default, `requireMention` is `true` and should be left that way for QQ.
+
+See [Group Chats](./overview#group-chats) for full details on group policies and mention gating.
+
+## Markdown Support
+
+The QQ Bot channel supports Markdown formatting (`msg_type=2`). The agent's Markdown responses are sent as-is, and QQ renders them with rich formatting (bold, italic, code blocks, links, lists).
+
+If the QQ server rejects a Markdown message for any reason, the channel automatically retries it as plain text — so your messages always go through even if the bot's Markdown capability is restricted server-side.
+
+This is the opposite of the WeChat channel, which strips all Markdown. You can let the agent use full Markdown with the QQ channel.
+
+## Token Management
+
+Access tokens expire after approximately 2 hours. The channel automatically refreshes them at 80% of their TTL (typically ~1.6 hours). If a refresh fails, it retries after 60 seconds.
+
+Token refresh continues across WebSocket reconnects — the channel never goes offline due to an expired token as long as the AppID and AppSecret remain valid.
+
+## Connection Resilience
+
+- **Auto-reconnect:** On WebSocket disconnect, the channel retries with exponential backoff (up to 20 attempts, max 30 seconds between retries)
+- **Session resume:** If the WebSocket drops briefly, the channel uses QQ's `RESUME` opcode to restore the session without losing in-flight messages
+- **Cross-server context continuation:** Chat sessions and routing state are persisted to disk. If the daemon restarts, conversations continue from where they left off
+- **Heartbeat monitoring:** HEARTBEAT_ACK timeouts are detected and force a reconnection to avoid zombie connections
+- **Message deduplication:** Replayed messages after a reconnect are detected and skipped
+
+## Tips
+
+- **Use Markdown freely** — Unlike WeChat, QQ renders Markdown natively. Bold, code blocks, lists, and links all work.
+- **Keep responses under 2000 characters** — Longer responses are automatically split into chunks. Adding a length hint to your instructions helps the agent stay concise.
+- **Sandbox for testing** — Set `"sandbox": true` to use the sandbox API during development. No production messages will be affected.
+- **Restrict access** — Use `senderPolicy: "allowlist"` for a fixed set of QQ users, or `"pairing"` to approve new users from the CLI. See [DM Pairing](./overview#dm-pairing) for details.
+
+## Key Differences from Telegram
+
+| Area             | QQ Bot                                      | Telegram                                      |
+| ---------------- | ------------------------------------------- | --------------------------------------------- |
+| Authentication   | QR code login or AppID/AppSecret            | Static bot token from BotFather               |
+| Markdown         | Native QQ Markdown with plaintext fallback  | HTML-formatted from agent Markdown            |
+| Token lifecycle  | 2h TTL, auto-refresh at 80%                 | Permanent bot token                           |
+| Group messages   | Only @mention messages are delivered to bot | Bot sees all messages (with privacy mode off) |
+| Typing indicator | Not available (QQ API limitation)           | "Working..." message                          |
+| Sandbox mode     | Supported for testing                       | Not available                                 |
+
+## Troubleshooting
+
+### Bot doesn't respond
+
+- Check the terminal output for errors
+- Verify the channel is running (`qwen channel status`)
+- If using `senderPolicy: "allowlist"`, make sure your QQ user ID is in `allowedUsers`
+- On first start, a QR code will appear in the terminal — scan it with your QQ app
+
+### Bot doesn't respond in groups
+
+- Check that `groupPolicy` is set to `"allowlist"` or `"open"` (default is `"disabled"`)
+- **You must @mention the bot** — QQ only delivers messages that tag the bot
+- Verify the bot has been added to the group
+
+### QR code login is stuck
+
+- The QR code is displayed in the terminal. Scan it with your QQ mobile app (Me → Scan)
+- If the QR code expires (typically after a few minutes), restart the channel to get a new one
+
+### Markdown messages appear as plain text
+
+- The QQ server may have rejected the Markdown message and the channel silently fell back to plain text. Check the terminal for `"Markdown rejected"` log messages
+- This is unusual on the QQ Bot Open Platform but can happen if the bot's Markdown capability is restricted server-side
+
+### Token expired after long downtime
+
+- If the channel is offline for more than 2 hours, the access token will have expired. The channel fetches a fresh token on reconnect — no action needed
+- If the AppSecret itself is invalid (e.g., rotated in the developer portal), update the `appSecret` field or delete `~/.qwen/channels/<name>-credentials.json` to re-trigger QR code login