weflow-cli 1.15.0 → 1.16.0

package/CHANGELOG.md

@@ -17,7 +17,23 @@ purpose** - a hand-written "latest is X" is exactly the claim that goes stale he
 
 All notable user-facing changes are recorded here. This project follows [Semantic Versioning](https://semver.org/).
 
-## 1.15.0
+## 1.16.0
+
+- **`leads` - the offers and asks buried in chat, with the sentence that said them.** Scans recent
+  conversations for five signal families (contact details, price, purchase, partnership, a request)
+  and ranks them, because the ones worth answering combine "how much" with "here is my number". Every
+  hit keeps the original words and whether you have replied yet (没回 / 我回过 / 已跟进 / 已结束).
+  Rule-based and offline on purpose: the output is a shortlist to read, not a verdict.
+  Two filters had to exist before it was usable - the first version returned eight "leads" in thirty
+  days and **every one was a campus-recruiting poster or a shopping promo**, because those are exactly
+  the messages salted with 合作 / 采购 / 报价. Adverts are now dropped, a signal must sit inside a
+  question or a request, and `@mentions` plus URLs are stripped before matching (group nicknames often
+  contain a phone number, and a URL with a long numeric id reads as one).
+- **`bonds` now shows the last thing said in a cooling conversation.** "已静默 304 天" does not tell
+  you whether to reach out; the final line does. That was the missing half of a "who should I contact"
+  reminder - the ranking and the silence were already computed.
+
+## 1.15.1
 
 - **Four read-only answers to questions WeChat cannot answer.** `weflow-cli today` shows what you were
   talking about on this day in earlier years (it walks every shard and every conversation table, so

package/OPERATIONS.md

@@ -1095,7 +1095,16 @@ powershell -NoProfile -ExecutionPolicy Bypass -File <仓库>\scripts\install_tas
 下游（AI 报告、统计）静默少一半。
 
 动作只有一句"跑那个脚本"，具体的事在仓库里的 `scripts/daily_task.ps1`（纯 ASCII、路径全部相对自身）。
-第一个位置参数是模式（`noon` = 今天；缺省 = 昨天）。它做四件事，每件都有理由：
+第一个位置参数是模式（`noon` = 今天；缺省 = 昨天）。
+
+> **动作是 `wscript.exe //B //Nologo run_hidden.vbs [noon]`，不是直接跑 PowerShell。**
+> `powershell.exe` 是**控制台程序**：计划任务一触发，Windows 会先给它建控制台窗口，之后 PowerShell
+> 才轮到隐藏自己 —— 结果就是每次运行都闪出一个黑框，全屏游戏时更明显（2026-10-22 实测：00:05 那次
+> 弹窗被人顺手关掉，进程以 `0xC000013A`（Ctrl+C 类中断）退出，那天就没有日报也没有邮件）。
+> `wscript.exe` 是没有控制台的 GUI 进程，`run_hidden.vbs` 内部再用 `WshShell.Run(cmd, 0, True)`
+> 隐藏子进程并等待它结束 —— 全程零窗口，而 `True` 让任务的"上次运行结果"仍然是真实退出码。
+
+它做五件事，每件都有理由：
 
 1. **跑目标天**：日终 `pipeline.py --yesterday --prefetch-bodies`；午报 `pipeline.py --date <今天> --prefetch-bodies`。
 2. **把输出落盘**：`logs/daily-<日期>[-noon].log` 与 `daily-<日期>[-noon].err.log`（`logs/` 已 gitignore）。
@@ -1110,6 +1119,12 @@ powershell -NoProfile -ExecutionPolicy Bypass -File <仓库>\scripts\install_tas
    午报也会写 index.html，只看它会让 00:05 跳过下午、留下一份"看着完整、实际半天"的日报。
    实际判据是 `.articles.json` 的 `generated_at`：**生成于目标日之后**（通常就是次日 00:0x），
    或当晚 23:00 之后，才算这天做完了。
+5. **渲染失败单独重试**：`pipeline.py` 的最后一环是渲染 `index.html`，而这一步已经挂过两次
+   （2026-10-18 无输出 11 小时；2026-10-23 在 README 之后中断，日报只剩 13:00 的半天版）。
+   抓取和摘要那时已经在 `.articles.json` 里，所以脚本会**单独**再跑一次 `generate_html.py`；
+   它成功就把这次运行当作成功，索引与邮件照常继续。判据是 `.articles.json` 比 `index.html` 新
+   （或页面不存在）——页面比数据新说明这一轮已经渲染过了，不重复。
+   **没有这一步，渲染慢一次就等于早上没有邮件**，因为发信挂在"整天成功"上。
 
 > 为什么不把 python 命令直接写进任务动作：① 实测 `Register-ScheduledTask` 对带嵌套引号的
 > `-Argument` 会报 `0x80004005`，而包一层脚本就不用跟引号搏斗；② 任务动作指向 CLI 那个 npm 包时，

package/README.en.md

@@ -19,7 +19,9 @@
 
 </div>
 
-**WeFlow CLI turns WeChat into your local second brain**: read and query chat history locally with one command, export readable HTML; turn official-account articles into AI-digested daily reports; read your WeChat favorites; build a semantic-searchable knowledge base — and plug it all into MCP-compatible AI editors like Claude Code. Works on Windows and Linux.
+**WeFlow CLI turns WeChat into your local second brain**: read and query chat history locally with one command, export readable HTML; turn official-account articles into AI-digested daily reports (optionally mailed to you); read your WeChat favorites; build a semantic-searchable knowledge base — voice messages included — and plug it all into MCP-compatible AI editors like Claude Code. Works on Windows and Linux.
+
+It is built to answer **questions WeChat itself cannot**: who is waiting on a reply, what I promised someone and whether I followed through, which relationships are cooling, how an account's position shifted over time, and what we were talking about on this day in earlier years.
 
 > 🔒 **Local-first**: runs entirely on your machine — zero telemetry, zero cloud reporting. Database keys are stored locally, encrypted with machine-bound AES-256-GCM. AI features only activate after you explicitly configure your own API key. See the [Security & Privacy statement](./SECURITY.md).
 
@@ -60,6 +62,17 @@ weflow-cli chat "How does RAG work?"   # knowledge-base RAG Q&A
 weflow-cli mcp-config                  # one-shot MCP client integration
 ```
 
+**Questions WeChat cannot answer** (all read-only, no network, no model calls):
+
+```text
+weflow-cli awaiting                    # who is waiting on a reply
+weflow-cli bonds                       # relationship temperature: cooling, warming, WeChat-only
+weflow-cli promises --days 180         # what I said I'd do, with the original words and follow-up evidence
+weflow-cli today --years 1,3,5         # what we were talking about on this day in earlier years
+weflow-cli network --with "contact"    # shared groups; --intro finds who could introduce you
+weflow-cli timeline --source "account" # one account's articles laid out in time
+```
+
 ## Platform Support
 
 | Capability | Windows | macOS | Linux |
@@ -283,6 +296,10 @@ Management: `weflow-cli assistant status` / `log` / `stop`; send "帮助" in WeC
 | Who is waiting on a reply | `weflow-cli awaiting --dry-run` · `weflow-cli awaiting --yes` |
 | Draft a reply (text unless you pick and send) | `weflow-cli draft <contact> --dry-run` · `--yes` (**only** `--pick N --send --yes` actually sends; both flags are required) |
 | Relationship temperature | `weflow-cli bonds` (who is cooling, who is warming, who only has WeChat; read-only, local, no model) |
+| What I promised whom | `weflow-cli promises --days 180` (original words plus follow-up evidence; rule-based, no model) |
+| This day in earlier years | `weflow-cli today --years 1,3,5` (what you were talking about back then) |
+| People graph | `weflow-cli network --groups` · `--with <name>` (shared groups) · `--intro <name>` (who could introduce you; reads `contact.db` only) |
+| One account over time | `weflow-cli timeline --list` · `--source <account>` (its articles laid out in time) |
 | Local judgement primitive | `weflow-cli decide --request <file> --dry-run` · `--yes` |
 | Export chat history | `weflow-cli export <contact> <json\|txt\|html\|excel>` |
 | Sync checkpoint | `weflow-cli sync run <contact> --since <date>` · `sync status` · `sync verify` |
@@ -290,7 +307,7 @@ Management: `weflow-cli assistant status` / `log` / `stop`; send "帮助" in WeC
 | Moments cache | `weflow-cli sns timeline` · `weflow-cli sns users` · `weflow-cli sns stats` |
 | WeChat favorites | `weflow-cli fav list` · `weflow-cli fav export markdown` · `weflow-cli fav set-key` |
 | WeRead | `weflow-cli weread shelf` · `notes` · `search` · `stats` (get the key from https://weread.qq.com/r/weread-skills; see [OPERATIONS.md](./OPERATIONS.md)) |
-| Knowledge base | `weflow-cli vault` · `weflow-cli wiki` · `weflow-cli search <query>` · `weflow-cli chat` (semantic search needs `config set dashscopeApiKey` first) |
+| Knowledge base | `weflow-cli vault` · `weflow-cli wiki` · `weflow-cli search <query>` · `weflow-cli chat` (semantic search needs `config set dashscopeApiKey` first; `search --phrase <words> --line all` also matches voice transcripts) |
 | Reports & tasks | `weflow-cli report` · `annual-report` · `todos` |
 | Second-brain assistant | `weflow-cli assistant start` · `status` · `log` · `stop` |
 | On-device panel | `weflow-cli panel` — floating window; `--status` reads state only, `--ask "…"` asks from the command line |
@@ -356,6 +373,8 @@ For a more complete view of modules and data flow, read [ARCHITECTURE.md](./ARCH
 - After upgrading WeChat, switching accounts, or migrating computers, you may need to re-initialize or re-scan NT keys.
 - `init` validates existing local configuration first and does not re-capture keys when that validation passes. Use `weflow-cli init --refresh` only after a migration, an account switch, or an access failure.
 - To test a first-time initialization or a failed key, prefer `weflow-cli init --test-missing-keys`; it simulates missing keys for that run only and leaves the saved configuration untouched. When keys really have failed, `weflow-cli config forget-keys` clears the database access keys only and asks for confirmation.
+- **Voice transcription runs on your machine** (`faster-whisper`, local model); audio never leaves it, and transcripts are cached by content hash so an interrupted run resumes. It is **not guaranteed accurate** — dialect, accent and background noise all introduce errors; the purpose is "you can find what was said", not a verbatim record. Transcripts join the phrase index only, never the semantic index (no second embedding bill for dialect-approximate text).
+- **Message shards are split by time** (`message_0..3.db`, roughly a year each), so "older chats" depend on whether the shard covering that period is still on the machine; `weflow-cli check --report` lists each shard's read status.
 - Running `weflow-cli daily` without a date checks yesterday's artifacts first: if yesterday's `README.md`, article index, or `index.html` is missing, it backfills yesterday and only generates today once that succeeds. Passing `--date` or `--dry-run` skips the backfill.
 
 ## Documentation & Feedback

package/README.md

@@ -19,7 +19,9 @@
 
 </div>
 
-**WeFlow CLI 把微信变成你的本地第二大脑**：在本机读取和查询聊天记录、导出可读 HTML；把公众号文章整理成带 AI 摘要的日报；读取微信收藏；构建可语义搜索的知识库；并通过 MCP 接入 Claude Code 等 AI 编辑器。支持 Windows 与 Linux。
+**WeFlow CLI 把微信变成你的本地第二大脑**：在本机读取和查询聊天记录、导出可读 HTML；把公众号文章整理成带 AI 摘要的日报（可邮件推送）；读取微信收藏；构建可语义搜索的知识库（连语音消息也能搜）；并通过 MCP 接入 Claude Code 等 AI 编辑器。支持 Windows 与 Linux。
+
+它专门回答**几个微信自己回答不了的问题**：谁在等我回话、我答应过谁什么后来做了没有、和谁的关系在变淡、某个号的观点怎么变的、那一年的今天我们在聊什么。
 
 > 🔒 **本地优先**：完全在你的电脑上运行——零遥测、零云上报；数据库密钥以机器绑定的 AES-256-GCM 加密存储在本机。AI 功能仅在显式配置你自己的 API Key 后启用。详见[安全与隐私声明](./SECURITY.md)。
 
@@ -60,6 +62,17 @@ weflow-cli chat "RAG 的原理是什么?"     # 知识库 RAG 问答
 weflow-cli mcp-config                  # 一键接入 Claude Code 等 MCP 客户端
 ```
 
+**几个微信给不出答案的问题**（都只读本地库，不联网、不调模型）：
+
+```text
+weflow-cli awaiting                    # 谁在等我回话
+weflow-cli bonds                       # 关系温度：谁在冷却、谁只有微信这一条线
+weflow-cli promises --days 180         # 我答应过谁什么，后来做了没有（带原话与后续证据）
+weflow-cli today --years 1,3,5         # 那几年的今天，我们在聊什么
+weflow-cli network --with "联系人A"     # 我和他共同在哪些群；--intro 看谁能引荐
+weflow-cli timeline --source "某号"     # 这个号历次说法的时间线
+```
+
 ## 平台支持
 
 | 能力 | Windows | macOS | Linux |
@@ -297,6 +310,7 @@ weflow-cli config set assistantGroupRequireMention true
 | 帮我起草回复（默认只出文字） | `weflow-cli draft <联系人> --dry-run` · `--yes`（**只有** `--pick N --send --yes` 会真的替你发出去；两步缺一不可）|
 | 关系温度 | `weflow-cli bonds`（谁在冷却、谁在变热、谁只有微信这一条线；只读本地库，不联网、不调模型）|
 | 我说过要给谁做什么 | `weflow-cli promises --days 180`（带回原话与后续证据，规则判定，不调模型）|
+| 聊天里的生意线索 | `weflow-cli leads --days 60 --min-score 4`（报价/采购/合作/需求信号，带原话与"我回过没有"）|
 | 那年今日 | `weflow-cli today --years 1,3,5`（几年前的今天在聊什么）|
 | 人脉图谱 | `weflow-cli network --groups` · `--with <名字>`（共同群）· `--intro <名字>`（谁能引荐；只读 contact.db）|
 | 某号观点演变 | `weflow-cli timeline --list` · `--source <公众号名>`（按时间看它说法的变化）|
@@ -307,7 +321,7 @@ weflow-cli config set assistantGroupRequireMention true
 | 朋友圈缓存 | `weflow-cli sns timeline` · `weflow-cli sns users` · `weflow-cli sns stats` |
 | 微信收藏 | `weflow-cli fav list` · `weflow-cli fav export markdown` · `weflow-cli fav set-key` |
 | 微信读书 | `weflow-cli weread shelf` · `notes` · `search` · `stats`（key 从 https://weread.qq.com/r/weread-skills 拿，见 [OPERATIONS.md](./OPERATIONS.md)） |
-| 知识库 | `weflow-cli vault` · `weflow-cli wiki` · `weflow-cli search <query>` · `weflow-cli chat`（语义检索需先 `config set dashscopeApiKey`）|
+| 知识库 | `weflow-cli vault` · `weflow-cli wiki` · `weflow-cli search <query>` · `weflow-cli chat`（语义检索需先 `config set dashscopeApiKey`；`search --phrase <词> --line all` 连语音转写一起搜）|
 | 总结与任务 | `weflow-cli report` · `annual-report` · `todos` |
 | 第二大脑助手 | `weflow-cli assistant start` · `status` · `log` · `stop` |
 | 本机面板 | `weflow-cli panel`（悬浮窗；`--status` 只看状态，`--ask "…"` 在命令行里问）|
@@ -374,6 +388,8 @@ python scripts/fav_server.py --date YYYY-MM-DD
 - `init` 会先验证已有本地配置；验证通过时不再重复捕获密钥。仅在迁移、切换账号或访问失败后使用 `weflow-cli init --refresh`。
 - 测试首次初始化或密钥失效时，优先使用 `weflow-cli init --test-missing-keys`；它只在本次运行模拟密钥缺失，不改动已保存配置。实际密钥失效时可使用 `weflow-cli config forget-keys`，该命令仅清除数据库访问密钥并要求确认。
 - 不指定日期运行 `weflow-cli daily` 时，会先检查昨天的日报产物；昨天缺少 `README.md`、文章索引或 `index.html` 时，会先补齐昨天，成功后才生成今天。指定 `--date` 或使用 `--dry-run` 时不执行补日报。
+- 语音转写用**本机** Whisper 模型（`faster-whisper`），音频不出本机；转写按内容哈希缓存，可中断续跑。它**不保证准确**——方言、口音和背景音都会带来误差，用途是"能搜到说过什么"，不是逐字记录。转写结果只并入短语索引，不进语义索引（不为方言近似文本再付一次嵌入费用）。
+- 消息分片按**时间段**切分（`message_0..3.db` 各覆盖一年多），所以"更早的聊天"取决于那段时间对应的分片是否还在本机；`weflow-cli check --report` 会列出每个分片的读取状态。
 
 ## 文档与反馈
 

package/bin/weflow-cli.ts

@@ -7561,6 +7561,47 @@ program
     }
   })
 
+// ==================== leads（线索雷达）====================
+program
+  .command('leads')
+  .description('线索雷达：聊天里的报价/采购/合作/需求信号，带原话与跟进状态（规则判定，不调模型）')
+  .option('--days <number>', '看最近多少天', '30')
+  .option('--min-score <number>', '最低分，4 分以上更接近真实生意', '2')
+  .option('--limit <number>', '最多列几条', '20')
+  .option('--json', '输出 JSON 格式')
+  .action(async (opts) => {
+    const { execFile } = await import('child_process')
+    const { promisify } = await import('util')
+    const execFileAsync = promisify(execFile)
+    const script = join(resolvePackageRoot(), 'scripts', 'leads.py')
+    const days = parseCliInteger(opts.days, 'days', 1, 3650, opts.json)
+    const minScore = parseCliInteger(opts.minScore, 'min-score', 1, 50, opts.json)
+    const limit = parseCliInteger(opts.limit, 'limit', 1, 500, opts.json)
+
+    try {
+      const args = [script, '--days', String(days), '--min-score', String(minScore),
+                    '--limit', String(limit)]
+      if (opts.json) args.push('--json')
+      const { stdout } = await execFileAsync(getPythonCommand(), args, {
+        timeout: 300_000,
+        maxBuffer: 10 * 1024 * 1024,
+        env: pythonProcessEnv(),
+      })
+      console.log(stdout)
+    } catch (e: any) {
+      if (opts.json) {
+        console.log(JSON.stringify({
+          success: false,
+          code: 'LEADS_FAILED',
+          error: '线索雷达不可用，请运行 check --json 检查配置和依赖',
+        }))
+        process.exit(1)
+      }
+      console.error(chalk.red(`\n${safeSubprocessError(e, '统计失败')}`))
+      process.exit(1)
+    }
+  })
+
 // ==================== timeline（某号观点演变）====================
 program
   .command('timeline')

package/dist/bin/weflow-cli.js

@@ -7849,6 +7849,47 @@ program
         process.exit(1);
     }
 });
+// ==================== leads（线索雷达）====================
+program
+    .command('leads')
+    .description('线索雷达：聊天里的报价/采购/合作/需求信号，带原话与跟进状态（规则判定，不调模型）')
+    .option('--days <number>', '看最近多少天', '30')
+    .option('--min-score <number>', '最低分，4 分以上更接近真实生意', '2')
+    .option('--limit <number>', '最多列几条', '20')
+    .option('--json', '输出 JSON 格式')
+    .action(async (opts) => {
+    const { execFile } = await import('child_process');
+    const { promisify } = await import('util');
+    const execFileAsync = promisify(execFile);
+    const script = join(resolvePackageRoot(), 'scripts', 'leads.py');
+    const days = parseCliInteger(opts.days, 'days', 1, 3650, opts.json);
+    const minScore = parseCliInteger(opts.minScore, 'min-score', 1, 50, opts.json);
+    const limit = parseCliInteger(opts.limit, 'limit', 1, 500, opts.json);
+    try {
+        const args = [script, '--days', String(days), '--min-score', String(minScore),
+            '--limit', String(limit)];
+        if (opts.json)
+            args.push('--json');
+        const { stdout } = await execFileAsync(getPythonCommand(), args, {
+            timeout: 300_000,
+            maxBuffer: 10 * 1024 * 1024,
+            env: pythonProcessEnv(),
+        });
+        console.log(stdout);
+    }
+    catch (e) {
+        if (opts.json) {
+            console.log(JSON.stringify({
+                success: false,
+                code: 'LEADS_FAILED',
+                error: '线索雷达不可用，请运行 check --json 检查配置和依赖',
+            }));
+            process.exit(1);
+        }
+        console.error(chalk.red(`\n${safeSubprocessError(e, '统计失败')}`));
+        process.exit(1);
+    }
+});
 // ==================== timeline（某号观点演变）====================
 program
     .command('timeline')