feat: add 9 new improvement items to Qwen Code report (26→35)#38
Conversation
New items from Claude Code source analysis: - P1: Commit Attribution (Co-Authored-By tracking) - P1: Conversation Branching (/branch fork sessions) - P1: Cost Tracking (USD + cache efficiency + per-model breakdown) - P2: Output Styles (Learning/Explanatory pedagogical modes) - P2: Fast Mode (speed/cost tier toggle) - P2: Concurrent Session Management (multi-terminal PID tracking) - P2: Git Diff Stats (structured diff + per-file statistics) - P2: File History Snapshots (per-file SHA256 backup vs git checkpoint) - P2: Session Ingress Auth (bearer token for remote sessions) Updated §二 matrix (35 items) and §四 summary table (25 dimensions). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
New articles: 1. cost-fastmode-deep-dive.md (166 lines) - Cost tracking: USD vs token counts, per-model breakdown, session persistence - Fast Mode: speed/cost toggle with cooldown + retry integration - Concurrent sessions: PID tracking + background agent detach 2. git-workflow-session-deep-dive.md (215 lines) - Commit Attribution: Co-Authored-By + per-file character contribution - Git Diff: native numstat/hunks parsing with file/size limits - File History: per-file SHA256 snapshots (100/session) vs git worktree - Conversation Branching: /branch with forkedFrom lineage - Output Styles: Learning (TODO(human)) + Explanatory modes Updated improvement report: - 8 of 9 new matrix items now link to deep-dive articles - §五 expanded with 2 new article entries (15→17 comparison articles) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
This PR expands the Qwen Code improvement report with 9 additional improvement items (26 → 35) and adds two new deep-dive documents to substantiate the new dimensions around Git workflow/session management and cost/Fast Mode comparisons.
Changes:
- Added 9 new improvement items to the Qwen Code improvement matrix and architecture-difference summary.
- Added two new deep-dive docs: one focused on Git workflow + session management, and one on cost tracking + Fast Mode + concurrent sessions.
- Updated the “相关 Deep-Dive 文章” index to link to the newly added deep-dive docs.
Reviewed changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
| docs/comparison/qwen-code-improvement-report.md | Adds the 9 new improvement items and updates the deep-dive index links. |
| docs/comparison/git-workflow-session-deep-dive.md | New deep-dive documenting commit attribution, diff stats, file history snapshots, session branching, and output styles. |
| docs/comparison/cost-fastmode-deep-dive.md | New deep-dive documenting USD cost tracking, Fast Mode, and concurrent session management vs Qwen Code’s stats/fast model behavior. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| | **P2** | FileIndex（fzf 风格模糊搜索） | 依赖 rg/glob | 中 | — | | ||
| | **P2** | ConfigTool（工具化设置读写） | 仅 /settings 命令 | 小 | — | | ||
| | **P2** | 自动后台化 Agent（超时转后台） | 需显式指定 | 小 | — | | ||
| | **P1** | [Commit Attribution](./git-workflow-session-deep-dive.md)（Co-Authored-By 追踪） | 缺失 | 小 | — | | ||
| | **P1** | [会话分支](./git-workflow-session-deep-dive.md)（/branch 对话分叉） | 缺失 | 中 | — | |
There was a problem hiding this comment.
The improvement matrix is currently grouped by priority (P0 → P1 → P2 → P3), but these newly added P1 items are appended after several P2 rows. This makes the table harder to scan and breaks the existing ordering—please move the new P1 rows up into the P1 block (before the first P2 row).
| |------|------------|-----------| | ||
| | **成本显示** | USD 金额 + 按模型分项 + cache 效率 | Token 计数 + 请求数 + cache 效率百分比 | | ||
| | **成本持久化** | 按 session ID 存储在项目配置中 | 无持久化 | | ||
| | **Fast Mode** | ✅ Opus 4.6 标准/快速切换（$5→$30/Mtok） | ❌（仅 `--fast` 指定备用模型） | |
There was a problem hiding this comment.
In the overview table, Fast Mode pricing is summarized as “$5→$30/Mtok”, but later in the same document you describe separate input/output pricing (“$5/$25 → $30/$150 per Mtok”). Please make the overview row consistent (e.g., include both rates or use a neutral wording like “显著更贵”).
| | **Fast Mode** | ✅ Opus 4.6 标准/快速切换（$5→$30/Mtok） | ❌（仅 `--fast` 指定备用模型） | | |
| | **Fast Mode** | ✅ Opus 4.6 标准/快速切换（$5/$25 → $30/$150 per Mtok） | ❌（仅 `--fast` 指定备用模型） | |
New items from deeper Claude Code module exploration: - P2: Computer Use (macOS desktop automation via Swift) - P2: LSP Tool (go-to-definition, find-references, symbol search) - P2: Deep Link protocol (claude-cli:// URI scheme for IDE/browser) - P2: /btw side question (context-switch without history contamination) - P2: Notebook Edit (Jupyter cell editing + auto cell ID) - P2: Team Memory Sync (org-wide memory + gitleaks secret scanner) - P3: Feature Gates (GrowthBook remote feature flags + A/B testing) - P3: DXT/MCPB bundle format (zip bomb protection + size limits) Updated §四 summary table with 7 new dimensions (25→32). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Final module sweep findings: - P2: Custom keybindings (multi-chord + platform adaptation + keybindings.json) - P2: Terminal theme detection (OSC 11 dark/light + COLORFGBG fallback) - P2: Corporate proxy support (CONNECT relay + CA cert injection + NO_PROXY) - P3: Memory diagnostics (V8 heap dump + 1.5GB threshold + leak detection) Updated §四 summary table (32→35 dimensions). Total items: 47 (P0×3, P1×11, P2×22, P3×11). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 3 out of 3 changed files in this pull request and generated 5 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| | **P2** | Computer Use（macOS 桌面自动化：截图 + 鼠标/键盘 + 剪贴板） | 缺失 | 大 | — | | ||
| | **P2** | LSP Tool（go-to-definition / find-references / symbol search） | 缺失 | 中 | — | | ||
| | **P2** | Deep Link 协议（`claude-cli://` URI scheme，IDE/浏览器集成） | 缺失 | 中 | — | | ||
| | **P2** | /btw 侧问（不中断主对话的快速提问） | 缺失 | 小 | — | | ||
| | **P2** | Notebook Edit（Jupyter cell 编辑 + 自动 cell ID 追踪） | 缺失 | 中 | — | |
There was a problem hiding this comment.
PR 描述/标题写的是“新增 9 个改进点（26→35）”，但这里新增了额外的 P2 条目（Computer Use、LSP Tool、Deep Link、/btw、Notebook Edit 等），会把总条目数推高到 35 以上。请确认这些是否也应计入本次 PR：要么更新 PR 描述/计数，要么把这些额外条目移到另一个 PR。
| | **P2** | Team Memory Sync（组织级记忆同步 + gitleaks 密钥扫描） | 缺失 | 大 | — | | ||
| | **P2** | 自定义快捷键（multi-chord + 平台适配 + `keybindings.json`） | 缺失 | 中 | — | | ||
| | **P2** | 终端主题检测（OSC 11 dark/light + COLORFGBG 回退） | 缺失 | 小 | — | |
There was a problem hiding this comment.
同上，这几项（Team Memory Sync / Feature Gates / DXT-MCPB）不在 PR 描述列出的 9 个新增条目里，会导致“26→35”的计数不成立。请确认是否为误加或需要同步更新 PR 描述与各处统计口径。
| | 文件历史快照 | per-file SHA256 + 按消息恢复 | checkpoint（git 级） | 小差距 | — | | ||
| | Session Ingress Auth | bearer token 远程认证 | 无 | 缺失 | — | | ||
| | Computer Use | macOS 桌面自动化 | 无 | 缺失 | — | | ||
| | LSP Tool | 代码智能导航 | 无 | 缺失 | — | |
There was a problem hiding this comment.
命名不一致：上面的改进矩阵用的是“Team Memory Sync”，但这里汇总维度写成“Team Memory”。建议统一命名（以及是否需要保留/强调“Sync”与 gitleaks 扫描这两个子能力）。
| |------|------------|-----------| | ||
| | **成本显示** | USD 金额 + 按模型分项 + cache 效率 | Token 计数 + 请求数 + cache 效率百分比 | | ||
| | **成本持久化** | 按 session ID 存储在项目配置中 | 无持久化 | | ||
| | **Fast Mode** | ✅ Opus 4.6 标准/快速切换（$5→$30/Mtok） | ❌（仅 `--fast` 指定备用模型） | |
There was a problem hiding this comment.
同一文档内 Fast Mode 的定价表述不一致：这里写“$5→$30/Mtok”，但后文示例又写成“Standard $5/$25 → Fast $30/$150 per Mtok”。建议在总览表里明确是 input/output 两档价格，或用与后文一致的格式，避免读者误解。
| | **Fast Mode** | ✅ Opus 4.6 标准/快速切换（$5→$30/Mtok） | ❌（仅 `--fast` 指定备用模型） | | |
| | **Fast Mode** | ✅ Opus 4.6 标准/快速切换（Standard $5/$25 → Fast $30/$150 per Mtok） | ❌（仅 `--fast` 指定备用模型） | |
| | 并发 Session | 多终端 PID 追踪 + 后台脱附 | 无 | 缺失 | — | | ||
| | Git Diff 统计 | 结构化 diff + 按文件统计 | 无 git-aware stats | 中等差距 | — | | ||
| | 文件历史快照 | per-file SHA256 + 按消息恢复 | checkpoint（git 级） | 小差距 | — | | ||
| | Session Ingress Auth | bearer token 远程认证 | 无 | 缺失 | — | | ||
| | Computer Use | macOS 桌面自动化 | 无 | 缺失 | — | | ||
| | LSP Tool | 代码智能导航 | 无 | 缺失 | — | |
There was a problem hiding this comment.
PR 描述里“Summary table: 16 → 25 dimensions（+9）”与这里新增的维度数量不一致：除 9 个条目外，这里还新增了 Computer Use/LSP Tool/Deep Link//btw/Notebook Edit/Team Memory 等维度。建议对齐 PR 描述的维度增量口径，避免读者以为只有 +9。
15 new detailed sections (6-20), each with: Claude Code implementation, source references, Qwen Code status, 缺失后果, 改进收益, related articles, and implementation suggestions. New items: 6. Mid-Turn Queue Drain (P0) — PR #2854 open 7. Kind-based tool parallelism (P1) — PR #2864 open 8. Startup optimization (P1) — preconnect + early input 9. Instruction conditional rules (P1) — frontmatter paths + lazy load 10. History Snip context collapse (P1) 11. ToolSearchTool dynamic discovery (P1) 12. Commit Attribution Co-Authored-By (P1) 13. /branch conversation forking (P1) 14. Cost tracking USD + cache (P1) 15. Shell security enhancement (P2) — IFS/Unicode/Zsh 16. MDM enterprise config (P2) — plist/Registry/remote 17. API real-time token counting (P2) 18. Output Styles Learning/Explanatory (P2) 19. Fast Mode speed/cost tier (P2) 20. Computer Use desktop automation (P2) Total: 578 lines, 47 matrix items, 20 detailed sections. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| | **P1** | [Commit Attribution](./git-workflow-session-deep-dive.md)（Co-Authored-By 追踪） | 缺失 | 小 | — | | ||
| | **P1** | [会话分支](./git-workflow-session-deep-dive.md)（/branch 对话分叉） | 缺失 | 中 | — | | ||
| | **P1** | [成本追踪](./cost-fastmode-deep-dive.md)（USD + cache 效率 + 按模型分项） | 仅基础 token/请求计数 | 中 | — | |
There was a problem hiding this comment.
In the improvement matrix, these new P1 rows are appended after the existing P2 block (lines 51–59). The table is otherwise grouped by priority (P0 → P1 → P2 → P3), so this breaks scanability and makes the “P1 items” look incomplete. Please move the P1 entries up into the P1 section (before the first P2 row) or re-sort the matrix consistently by priority.
| | 维度 | Claude Code | Qwen Code | | ||
| |------|------------|-----------| | ||
| | **Commit Attribution** | ✅ Co-Authored-By + 按文件字符归因 + git notes | ❌ | | ||
| | **Git Diff 统计** | ✅ numstat + hunks 解析（50 文件 / 1MB 上限） | 依赖 simple-git npm | |
There was a problem hiding this comment.
The overview table suggests Qwen Code has “Git Diff 统计” via “依赖 simple-git npm”, but section 3.2 later states there’s no dedicated diff parser and no numstat fast probe/structured hunks output. Please align the overview row with the detailed section (e.g., mark it as missing/limited and briefly state what’s actually available) to avoid a self-contradiction.
| | **Git Diff 统计** | ✅ numstat + hunks 解析（50 文件 / 1MB 上限） | 依赖 simple-git npm | | |
| | **Git Diff 统计** | ✅ numstat + hunks 解析（50 文件 / 1MB 上限） | ⚠️ 有限：依赖 simple-git 获取基础输出，无专用 diff parser / numstat fast probe / structured hunks | |
Removed "成本追踪 USD + cache 效率" (P1) from matrix, detailed section, and summary table. Qwen Code supports multiple providers with different pricing — USD cost calculation is not directly applicable. Renumbered sections 15-20 → 14-19. Title: Top 20 → Top 19. Kept cost-fastmode-deep-dive.md link in §五 (still has Fast Mode content). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Added P2 item: Denial Tracking (权限拒绝学习 + 自动回退到 prompting) - Claude Code: utils/permissions/denialTracking.ts (45 lines) - Records consecutive denials, auto-fallback to prompting when threshold hit - Prevents classifier from silently blocking legitimate operations Restored title to "Top 20". Matrix: 47 items, detailed sections: 20. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| | **P1** | [Commit Attribution](./git-workflow-session-deep-dive.md)（Co-Authored-By 追踪） | 缺失 | 小 | — | | ||
| | **P1** | [会话分支](./git-workflow-session-deep-dive.md)（/branch 对话分叉） | 缺失 | 中 | — | | ||
| | **P2** | [Output Styles](./git-workflow-session-deep-dive.md)（Learning 教学模式 / Explanatory 解释模式） | 缺失 | 中 | — | | ||
| | **P2** | [Fast Mode](./cost-fastmode-deep-dive.md)（速度/成本分级推理） | 缺失 | 小 | — | | ||
| | **P2** | [并发 Session](./cost-fastmode-deep-dive.md) 管理（多终端 PID 追踪 + 后台 Agent 脱附） | 缺失 | 中 | — | | ||
| | **P2** | [Git Diff 统计](./git-workflow-session-deep-dive.md)（工具编辑后结构化 diff + 按文件统计） | 无 git-aware diff stats | 小 | — | | ||
| | **P2** | [文件历史快照](./git-workflow-session-deep-dive.md)（per-file SHA256 备份 + 按消息恢复） | checkpoint 模式（git 级） | 中 | — | | ||
| | **P2** | Session Ingress Auth（远程会话 bearer token 认证） | 缺失 | 中 | — | | ||
| | **P2** | Computer Use（macOS 桌面自动化：截图 + 鼠标/键盘 + 剪贴板） | 缺失 | 大 | — | | ||
| | **P2** | LSP Tool（go-to-definition / find-references / symbol search） | 缺失 | 中 | — | | ||
| | **P2** | Deep Link 协议（`claude-cli://` URI scheme，IDE/浏览器集成） | 缺失 | 中 | — | | ||
| | **P2** | /btw 侧问（不中断主对话的快速提问） | 缺失 | 小 | — | | ||
| | **P2** | Notebook Edit（Jupyter cell 编辑 + 自动 cell ID 追踪） | 缺失 | 中 | — | | ||
| | **P2** | Team Memory Sync（组织级记忆同步 + gitleaks 密钥扫描） | 缺失 | 大 | — | | ||
| | **P2** | 自定义快捷键（multi-chord + 平台适配 + `keybindings.json`） | 缺失 | 中 | — | | ||
| | **P2** | 终端主题检测（OSC 11 dark/light + COLORFGBG 回退） | 缺失 | 小 | — | | ||
| | **P2** | 企业代理支持（CONNECT relay + CA cert 注入 + NO_PROXY 白名单） | 缺失 | 大 | — | |
There was a problem hiding this comment.
PR 标题/描述里写的是“新增 9 个改进项（26→35）”，但该矩阵在此处一次性新增了多条改进项（从 Commit Attribution 一直新增到 企业代理支持 等）。请对齐 PR 元数据与文档内容：要么更新 PR 标题/描述中的数量与分布统计，要么将本次不打算纳入的条目移出该报告/拆分到其他 PR，避免读者误解变更范围。
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…mentation Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| | **P1** | [Commit Attribution](./git-workflow-session-deep-dive.md)（Co-Authored-By 追踪） | 缺失 | 小 | — | | ||
| | **P1** | [会话分支](./git-workflow-session-deep-dive.md)（/branch 对话分叉） | 缺失 | 中 | — | | ||
| | **P2** | [Output Styles](./git-workflow-session-deep-dive.md)（Learning 教学模式 / Explanatory 解释模式） | 缺失 | 中 | — | | ||
| | **P2** | [Fast Mode](./cost-fastmode-deep-dive.md)（速度/成本分级推理） | 缺失 | 小 | — | | ||
| | **P2** | [并发 Session](./cost-fastmode-deep-dive.md) 管理（多终端 PID 追踪 + 后台 Agent 脱附） | 缺失 | 中 | — | |
There was a problem hiding this comment.
PR 标题/描述写的是“新增 9 个改进项（26→35）”，但该矩阵在此处开始新增的条目从 60 行一直延伸到 79 行（共 20 条新增行），与“+9”不一致。建议同步修正 PR 描述/标题或调整文档中新增条目数量，避免读者误解。
New articles: 1. computer-use-deep-dive.md (131 lines) — macOS desktop automation via MCP: SCContentFilter screenshots, enigo mouse/keyboard, TCC permissions 2. deep-link-protocol-deep-dive.md (156 lines) — claude-cli:// URI scheme: terminal auto-detect (10+ terminals × 3 platforms), GitHub repo resolve 3. tool-search-deep-dive.md (131 lines) — ToolSearchTool deferred loading: keyword/select search, MCP-weighted scoring, lazy schema loading Updated improvement report: - All 13 P0+P1 items now have deep-dive article links (was 10/13) - Fixed History Snip description: "实现" → "scaffolding 阶段" - Added 3 new articles to §五 index (17→20 comparison articles) 4-round audit: markdown OK, terms clean, files exist, 13/13 P0/P1 linked. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 6 out of 6 changed files in this pull request and generated 2 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| | **P2** | FileIndex（fzf 风格模糊搜索） | 依赖 rg/glob | 中 | — | | ||
| | **P2** | ConfigTool（工具化设置读写） | 仅 /settings 命令 | 小 | — | | ||
| | **P2** | 自动后台化 Agent（超时转后台） | 需显式指定 | 小 | — | | ||
| | **P1** | [Commit Attribution](./git-workflow-session-deep-dive.md)（Co-Authored-By 追踪） | 缺失 | 小 | — | | ||
| | **P1** | [会话分支](./git-workflow-session-deep-dive.md)（/branch 对话分叉） | 缺失 | 中 | — | | ||
| | **P2** | [Output Styles](./git-workflow-session-deep-dive.md)（Learning 教学模式 / Explanatory 解释模式） | 缺失 | 中 | — | | ||
| | **P2** | [Fast Mode](./cost-fastmode-deep-dive.md)（速度/成本分级推理） | 缺失 | 小 | — | | ||
| | **P2** | [并发 Session](./cost-fastmode-deep-dive.md) 管理（多终端 PID 追踪 + 后台 Agent 脱附） | 缺失 | 中 | — | | ||
| | **P2** | [Git Diff 统计](./git-workflow-session-deep-dive.md)（工具编辑后结构化 diff + 按文件统计） | 无 git-aware diff stats | 小 | — | | ||
| | **P2** | [文件历史快照](./git-workflow-session-deep-dive.md)（per-file SHA256 备份 + 按消息恢复） | checkpoint 模式（git 级） | 中 | — | |
There was a problem hiding this comment.
改进建议矩阵的优先级排序在这里出现了“P2 条目之后又插入 P1 条目”的情况（Commit Attribution、会话分支）。这会降低表格可读性，也会让后续统计/引用优先级时更容易出错。建议按 P0→P1→P2→P3 分组/排序（或增加分隔小标题），并将新增 P1 条目移动到 P1 区块中。
|
|
||
| ```typescript | ||
| // 源码: tools/ToolSearchTool/prompt.ts#L62-L108 | ||
| // 分类逻辑: |
There was a problem hiding this comment.
这里的 TypeScript 代码块目前只有注释、没有任何示例代码（空 code fence）。在渲染时会显得像是内容缺失。建议：要么补充一段最小的分类伪代码/片段来说明 alwaysLoad/shouldDefer 的判定逻辑，要么移除 code fence，改用普通段落描述并保留源码引用。
| // 分类逻辑: | |
| function classifyTool(tool: { | |
| name: string; | |
| alwaysLoad?: boolean; | |
| shouldDefer?: boolean; | |
| }) { | |
| if (tool.name.startsWith("mcp__")) return "deferred"; | |
| if (tool.alwaysLoad) return "always-load"; | |
| if (tool.shouldDefer) return "deferred"; | |
| return "always-load"; | |
| } |
New article covering org-wide memory sync: - API protocol: GET/PUT with ETag + SHA256 per-key checksums - Delta upload: only changed keys, 200KB batch limit - Conflict resolution: 412 → refresh hashes → retry (max 2) - Gitleaks secret scanning: 29 rules, client-side before upload - File watching: fs.watch + 2s debounce push - Symlink security (PSR M22186) Updated improvement report: - Team Memory Sync P2 item now linked - §五 index expanded with team-memory-deep-dive.md Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 10 out of 10 changed files in this pull request and generated no new comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Each of 22 items now has: - 思路: 2-3 sentence approach overview - Claude Code 源码索引: file + key function/constant table - Qwen Code 修改方向: where to modify, one sentence - 相关文章 link Format optimized for programmers: grep the function name → find the Claude Code source → understand the approach → implement in Qwen Code. Fixed table column mismatch (pipe in HookConfig type). 433 lines, 23 anchors, 14 deep-dive links (all valid). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 10 out of 10 changed files in this pull request and generated 1 comment.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
|
|
||
| | 文件 | 内容 | 项数 | | ||
| |------|------|:----:| | ||
| | [P0/P1 详细说明](./qwen-code-improvement-report-p0-p1.md) | 最高优先级（Mid-Turn Drain、压缩、Fork、记忆、并行等） | 22 | |
There was a problem hiding this comment.
The item count for the P0/P1 detail doc appears incorrect. qwen-code-improvement-report-p0-p1.md defines anchors for items 1–13, 49–52, and 58–63 (23 items total), but the summary table lists 22. Please reconcile the count (or adjust the included items) so the table stays accurate.
| | [P0/P1 详细说明](./qwen-code-improvement-report-p0-p1.md) | 最高优先级（Mid-Turn Drain、压缩、Fork、记忆、并行等） | 22 | | |
| | [P0/P1 详细说明](./qwen-code-improvement-report-p0-p1.md) | 最高优先级（Mid-Turn Drain、压缩、Fork、记忆、并行等） | 23 | |
…links All 3 detail files (P0/P1, P2, P3) rewritten with consistent format: - 思路: 2-3 sentence approach overview - Claude Code 源码索引: file + key function/constant table - Qwen Code 修改方向: where to modify, one sentence Main file anchor links fixed: [↓](#item-N) → [↓](./sub-file.md#item-N) to correctly point across files after split. Validation: 70 links = 70 anchors, 0 broken, 0 table issues. Files: main 198L + P0/P1 433L + P2 626L + P3 170L = 1,427L total. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 10 out of 10 changed files in this pull request and generated 1 comment.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| | [P0/P1 详细说明](./qwen-code-improvement-report-p0-p1.md) | 最高优先级（Mid-Turn Drain、压缩、Fork、记忆、并行等） | 22 | | ||
| | [P2 详细说明](./qwen-code-improvement-report-p2.md) | 中等优先级（Shell 安全、MDM、Computer Use、Deep Link 等） | 37 | |
There was a problem hiding this comment.
“全部改进点详细说明”这里的“项数”统计与实际条目数不一致：
qwen-code-improvement-report-p0-p1.md中共有 23 个item-*锚点（1-13、49-52、58-63），此处写为 22。qwen-code-improvement-report-p2.md中共有 36 个item-*锚点（14-36、48、53-57、64、66-71），此处写为 37。
建议根据各文件实际条目数更新统计，避免读者误解总量。
| | [P0/P1 详细说明](./qwen-code-improvement-report-p0-p1.md) | 最高优先级（Mid-Turn Drain、压缩、Fork、记忆、并行等） | 22 | | |
| | [P2 详细说明](./qwen-code-improvement-report-p2.md) | 中等优先级（Shell 安全、MDM、Computer Use、Deep Link 等） | 37 | | |
| | [P0/P1 详细说明](./qwen-code-improvement-report-p0-p1.md) | 最高优先级（Mid-Turn Drain、压缩、Fork、记忆、并行等） | 23 | | |
| | [P2 详细说明](./qwen-code-improvement-report-p2.md) | 中等优先级（Shell 安全、MDM、Computer Use、Deep Link 等） | 36 | |
Every item now has 3 decision-support lines: - 意义: why this feature matters (1 sentence) - 缺失后果: what happens without it (1 sentence) - 改进收益: what you gain (1 sentence) Helps decision-makers quickly judge "值不值得做" for each item. Also fixed all 70 anchor links: [↓](#item-N) → [↓](./sub-file.md#item-N) to correctly point across files after split. Files: P0/P1 521L + P2 766L + P3 210L + main 198L = 1,695L total. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 10 out of 10 changed files in this pull request and generated 3 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| | [P0/P1 详细说明](./qwen-code-improvement-report-p0-p1.md) | 最高优先级（Mid-Turn Drain、压缩、Fork、记忆、并行等） | 22 | | ||
| | [P2 详细说明](./qwen-code-improvement-report-p2.md) | 中等优先级（Shell 安全、MDM、Computer Use、Deep Link 等） | 37 | | ||
| | [P3 详细说明](./qwen-code-improvement-report-p3.md) | 低优先级（Feature Gates、Vim、语音、插件市场等） | 11 | |
There was a problem hiding this comment.
The per-priority item counts here sum to 70 items (22 + 37 + 11), which doesn’t match the PR title/description claiming the report goes from 26 → 35 items. Please reconcile the numbers (either update this table + any related totals, or adjust the PR title/description to reflect the expanded scope).
| | **P2** | [文件历史快照](./git-workflow-session-deep-dive.md) — per-file SHA256 备份，按消息粒度恢复（100 个/session） [↓](./qwen-code-improvement-report-p2.md#item-23) | git-level checkpoint | 中 | — | | ||
| | **P2** | [Computer Use](./computer-use-deep-dive.md) — macOS 截图 + 鼠标/键盘 + 剪贴板，通过 MCP 桥接 [↓](./qwen-code-improvement-report-p2.md#item-19) | 缺失 | 大 | — | | ||
| | **P2** | [Deep Link](./deep-link-protocol-deep-dive.md) — `claude-cli://` 一键从浏览器/IDE 启动 Agent + 预填充 prompt [↓](./qwen-code-improvement-report-p2.md#item-24) | 缺失 | 中 | — | | ||
| | **P2** | [Team Memory](./team-memory-deep-dive.md) — 团队共享项目知识 + 29 条 gitleaks 密钥扫描 + ETag 同步 [↓](./qwen-code-improvement-report-p0-p1.md#item-10) | 缺失 | 大 | — | |
There was a problem hiding this comment.
The Team Memory row is marked as P2 in this matrix, but its detail link points to the P0/P1 detail file (qwen-code-improvement-report-p0-p1.md#item-10). This cross-file placement is easy to miss when navigating by priority. Consider moving the detailed section to the P2 doc (or adding a note in the matrix explaining why a P2 item’s details live under P0/P1).
| | **P2** | [Team Memory](./team-memory-deep-dive.md) — 团队共享项目知识 + 29 条 gitleaks 密钥扫描 + ETag 同步 [↓](./qwen-code-improvement-report-p0-p1.md#item-10) | 缺失 | 大 | — | | |
| | **P2** | [Team Memory](./team-memory-deep-dive.md) — 团队共享项目知识 + 29 条 gitleaks 密钥扫描 + ETag 同步（详情暂收录于 P0/P1 文档） [↓](./qwen-code-improvement-report-p0-p1.md#item-10) | 缺失 | 大 | — | |
| | **P3** | 语音模式 — push-to-talk 语音输入 + 流式 STT 转录 + 可重绑快捷键 [↓](./qwen-code-improvement-report-p3.md#item-46) | 缺失 | 大 | — | | ||
| | **P3** | [插件市场](./hook-plugin-extension-deep-dive.md) — 插件发现、安装、版本管理 + 前端 UI [↓](./qwen-code-improvement-report-p3.md#item-47) | 缺失 | 大 | — | | ||
|
|
||
| > 点击改进点名称可跳转到 Deep-Dive 文章；每项的详细说明（缺失后果 + 改进收益 + 建议方案）见 [§三](#三全部改进点详细说明)。 |
There was a problem hiding this comment.
This note says “点击改进点名称可跳转到 Deep-Dive 文章”, but several items in the “改进点” column are plain text (e.g., Plan 模式 Interview, BriefTool, FileIndex, Notebook Edit, 自定义快捷键, Session Ingress Auth, etc.) and therefore aren’t clickable. Either convert these titles into links (where applicable) or soften the wording to reflect that only some items have Deep-Dive pages.
| > 点击改进点名称可跳转到 Deep-Dive 文章；每项的详细说明（缺失后果 + 改进收益 + 建议方案）见 [§三](#三全部改进点详细说明)。 | |
| > 部分改进点名称可跳转到 Deep-Dive 文章；每项的详细说明（缺失后果 + 改进收益 + 建议方案）见 [§三](#三全部改进点详细说明)，也可点击各项后的 [↓] 跳转。 |
Qwen Code has: - commands/channel.ts (start/stop/status/pairing) - channel/ directory (registry, config, pairing, pidfile) - @qwen-code/channel-weixin (WeChat channel — Claude Code doesn't have) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 10 out of 10 changed files in this pull request and generated 1 comment.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| # Qwen Code 改进建议 — P0/P1 详细说明 | ||
|
|
||
| > 最高优先级改进项。每项包含：思路概述、Claude Code 源码索引（方便查找参考）、Qwen Code 修改方向。 |
There was a problem hiding this comment.
该文件标题/导语标注为“P0/P1 详细说明 / 最高优先级改进项”，但文内包含“Team Memory（P2→Top20）”等非 P0/P1 条目，容易让读者（以及总览页的按优先级索引）产生分类混淆。建议要么把非 P0/P1 条目移到对应优先级文件（如 P2），要么把文件命名与导语改为能覆盖 Top20/跨优先级的定位。
| # Qwen Code 改进建议 — P0/P1 详细说明 | |
| > 最高优先级改进项。每项包含：思路概述、Claude Code 源码索引（方便查找参考）、Qwen Code 修改方向。 | |
| # Qwen Code 改进建议 — Top20 / 跨优先级详细说明 | |
| > 按综合排序收录的重点改进项，可能包含 P0 / P1 以及进入 Top20 的其他优先级条目。每项包含：思路概述、Claude Code 源码索引（方便查找参考）、Qwen Code 修改方向。 |
After removing 5 misreports (btw, LSP, /loop, /context, Channels), re-numbered all sections: - P0/P1: items 1-22 (22 items) - P2: items 23-58 (36 items) - P3: items 59-69 (11 items) All 69 matrix links correctly point to sub-files. Removed duplicate --- separator from P0/P1 file. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 10 out of 10 changed files in this pull request and generated 3 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| <a id="item-41"></a> | ||
|
|
||
| ### 23. Shell 安全增强（P2） | ||
|
|
||
| **思路**：在 AST 读写分类基础上，补充专项检查——IFS 注入、Unicode 空白、Zsh 危险命令、花括号展开等。AST 是主路径（精确），专项检查是补充（覆盖面）。 |
There was a problem hiding this comment.
qwen-code-improvement-report.md links to anchors like qwen-code-improvement-report-p2.md#item-23, but this file’s anchors are misnumbered (e.g. item 23 is tagged as id="item-41") and some IDs are reused multiple times. This will break all intra-doc jumps into the P2 detail page. Please renumber the <a id="item-…"> blocks to match the section numbers (23–58) and ensure each ID is unique (or drop manual anchors and link to the generated heading anchors consistently).
| --- | ||
|
|
||
| <a id="item-37"></a> | ||
|
|
||
| ### 28. Computer Use 桌面自动化（P2） |
There was a problem hiding this comment.
The <a id="item-…"> anchors in this section don’t match the numbered headings (e.g. item 28 uses id="item-37", item 30 uses id="item-39"). Since the summary table links to specific #item-XX anchors, these mismatches will send readers to the wrong place or nowhere. Align each anchor ID with its corresponding item number and keep them monotonic/consistent throughout the file.
| | 文件 | 内容 | 项数 | | ||
| |------|------|:----:| | ||
| | [P0/P1 详细说明](./qwen-code-improvement-report-p0-p1.md) | 最高优先级（Mid-Turn Drain、压缩、Fork、记忆、并行等） | 22 | | ||
| | [P2 详细说明](./qwen-code-improvement-report-p2.md) | 中等优先级（Shell 安全、MDM、Computer Use、Deep Link 等） | 37 | |
There was a problem hiding this comment.
In §三 the per-file item counts appear incorrect: P2 detail items are numbered 23–58 (inclusive), which is 36 items, but the table shows 37. Please update the count to match the actual number of items (or adjust numbering/content so they’re consistent).
| | [P2 详细说明](./qwen-code-improvement-report-p2.md) | 中等优先级（Shell 安全、MDM、Computer Use、Deep Link 等） | 37 | | |
| | [P2 详细说明](./qwen-code-improvement-report-p2.md) | 中等优先级（Shell 安全、MDM、Computer Use、Deep Link 等） | 36 | |
- P2 file had duplicate/wrong anchors from rewrite — rebuilt all to match section numbers (item-23 through item-58) - Added missing 意义/缺失后果/改进收益 to sections 22, 58, 69 - Final: 69 anchors = 69 links = 69 sections = 69 impacts, 0 broken Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 10 out of 10 changed files in this pull request and generated no new comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Closed PR #75 — 10 of 16 items duplicated existing 187. Source code references all verified correct (glm-5.1 self-corrected 8 errors). Extracted 4 new items to p2-tools: - #35 preserveQuoteStyle (quote style preservation in edits) - #36 areFileEditsInputsEquivalent (skip duplicate edits) - #37 MCP channel plugin allowlist - #38 SDK message type enrichment (11 → 30+ types) Total: 191 items. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Claude Code's read_file supports PDF (poppler-utils), images (Vision), and Jupyter notebooks natively. Qwen Code currently rejects PDFs (#2024). Issue #38 ("支持二进制文件解析") has been open since project creation. References: MarkItDown (microsoft/markitdown, 102K stars) as MCP integration option for DOCX/XLSX/PPTX. 248 items (was 247), tools-commands 21 items (was 20). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Summary
Add 9 new improvement items discovered from further Claude Code source analysis:
Total items: 26 → 35 (P0×3, P1×11, P2×14, P3×7)
Summary table: 16 → 25 dimensions
Test plan
🤖 Generated with Claude Code