publishport-opencli 1.0.1 → 1.0.2

package/skills/opencli-adapter-author/references/coverage-matrix.md

@@ -39,7 +39,7 @@ skill 明确承诺能搞定什么、搞不定什么。动手前先看一眼这
 | | CSV 字符串（eastmoney kline） | ✅ | `response.split(',')` 按列序解 |
 | | HTML 表格（tonghuashun） | 🟡 | `page.evaluate` 里用 `querySelectorAll` 拿 |
 
-🟡 的维度意思：方法在文档里，但这一版没拿真实站点跑过端到端。第一次遇到时按文档走，遇到和文档不一致的地方记到 `~/.opencli/sites/<site>/notes.md`，下一次再打开就是 ✅。
+🟡 的维度意思：方法在文档里，但这一版没拿真实站点跑过端到端。第一次遇到时按文档走，遇到和文档不一致的地方记到 `~/.ppcli/sites/<site>/notes.md`，下一次再打开就是 ✅。
 
 ---
 
@@ -74,8 +74,8 @@ skill 明确承诺能搞定什么、搞不定什么。动手前先看一眼这
 
 | 类型 | 证据 adapter | 覆盖维度 |
 |------|-------------|---------|
-| PUBLIC + 自解释字段 + SPA | `~/.opencli/clis/coingecko/top.js`（本轮 dry run） | `Strategy.PUBLIC` + REST JSON + 自解释字段 + 列表页 |
+| PUBLIC + 自解释字段 + SPA | `~/.ppcli/clis/coingecko/top.js`（本轮 dry run） | `Strategy.PUBLIC` + REST JSON + 自解释字段 + 列表页 |
 | COOKIE + 代号字段 + JSONP | `clis/eastmoney/*.js` × 13（PR #1091 merged） | `Strategy.PUBLIC`（匿名 `ut=`）+ JSONP + f-代号 + 列表/详情/K 线 |
 | COOKIE + SPA | `clis/bilibili/*.js` × 10+（已存在） | `Strategy.COOKIE + browser:true` + wbi 签名 |
 
-**本 PR 新增的 skill 还未硬验证的维度**：🟡 行，尤其 SSR Pattern B + Bearer/CSRF + 未登记代号解码。这些放到第一批真实用户 adapter 写作中打磨，skill 文档先落，踩坑回来补 `site-memory`。合 PR 之前先拿 coingecko 跑第二轮（带着第一轮写出的 `~/.opencli/sites/coingecko/`）验证 memory 命中 → endpoint re-verify → 字段抽查 → 写 adapter 这条回路。
+**本 PR 新增的 skill 还未硬验证的维度**：🟡 行，尤其 SSR Pattern B + Bearer/CSRF + 未登记代号解码。这些放到第一批真实用户 adapter 写作中打磨，skill 文档先落，踩坑回来补 `site-memory`。合 PR 之前先拿 coingecko 跑第二轮（带着第一轮写出的 `~/.ppcli/sites/coingecko/`）验证 memory 命中 → endpoint re-verify → 字段抽查 → 写 adapter 这条回路。

package/skills/opencli-adapter-author/references/field-decode-playbook.md

@@ -42,7 +42,7 @@
 
 ```bash
 # 如果是相对路径，拼 domain 回浏览器验证
-opencli browser eval "window.location.origin + '/<path>'"
+ppcli browser eval "window.location.origin + '/<path>'"
 ```
 
 ---
@@ -68,8 +68,8 @@ opencli browser eval "window.location.origin + '/<path>'"
    比如 eastmoney，`fid=f2`（最新价）和 `fid=f3`（涨跌幅）：
 
    ```bash
-   opencli browser eval "fetch('<url>&fid=f2&po=1').then(r=>r.json()).then(d=>d.data.diff.slice(0,3))"
-   opencli browser eval "fetch('<url>&fid=f3&po=1').then(r=>r.json()).then(d=>d.data.diff.slice(0,3))"
+   ppcli browser eval "fetch('<url>&fid=f2&po=1').then(r=>r.json()).then(d=>d.data.diff.slice(0,3))"
+   ppcli browser eval "fetch('<url>&fid=f3&po=1').then(r=>r.json()).then(d=>d.data.diff.slice(0,3))"
    ```
 
 3. **对比两组数据**：
@@ -81,7 +81,7 @@ opencli browser eval "window.location.origin + '/<path>'"
 4. **用第三个参数交叉验证**
 
    ```bash
-   opencli browser eval "fetch('<url>&fid=f6&po=1').then(r=>r.json()).then(d=>d.data.diff.slice(0,3))"  # 成交额
+   ppcli browser eval "fetch('<url>&fid=f6&po=1').then(r=>r.json()).then(d=>d.data.diff.slice(0,3))"  # 成交额
    ```
 
    对照网页上"成交额排行"的前三名。对得上就认。
@@ -90,11 +90,11 @@ opencli browser eval "window.location.origin + '/<path>'"
 
 ```bash
 # 按价格排，拿一条观察
-opencli browser eval "fetch('https://push2.eastmoney.com/api/qt/clist/get?fs=b:MK0354&pn=1&pz=1&fid=f2&po=1&fltt=2&fields=f12,f14,f2,f3,f236,f237,f239').then(r=>r.json()).then(d=>d.data.diff[0])"
+ppcli browser eval "fetch('https://push2.eastmoney.com/api/qt/clist/get?fs=b:MK0354&pn=1&pz=1&fid=f2&po=1&fltt=2&fields=f12,f14,f2,f3,f236,f237,f239').then(r=>r.json()).then(d=>d.data.diff[0])"
 # 返回 {f12:'123456', f14:'XX转债', f2:180.5, f3:2.1, f236:98.5, f237:83.2, f239:-4.1}
 
 # 按 f237 排
-opencli browser eval "fetch('...&fid=f237&po=1...').then(...)"
+ppcli browser eval "fetch('...&fid=f237&po=1...').then(...)"
 # 第一条变了，f237 值变成 400+
 
 # 打开 eastmoney 可转债页，切"溢价率"排序，第一条的溢价率确实是 400+ —— f237 就是溢价率 %
@@ -111,13 +111,13 @@ opencli browser eval "fetch('...&fid=f237&po=1...').then(...)"
 1. **先数一次嵌套路径**
 
    ```bash
-   opencli browser eval "fetch('<url>').then(r=>r.json()).then(j=>({keys:Object.keys(j), type:Array.isArray(j)?'array':'object'}))"
+   ppcli browser eval "fetch('<url>').then(r=>r.json()).then(j=>({keys:Object.keys(j), type:Array.isArray(j)?'array':'object'}))"
    ```
 
 2. **一层一层剥**
 
    ```bash
-   opencli browser eval "fetch('<url>').then(r=>r.json()).then(j=>{const d=j.data; return {keys:Object.keys(d), sample: d[Object.keys(d)[0]]}})"
+   ppcli browser eval "fetch('<url>').then(r=>r.json()).then(j=>{const d=j.data; return {keys:Object.keys(d), sample: d[Object.keys(d)[0]]}})"
    ```
 
 3. **数数组长度对照 pz / pageSize**
@@ -128,7 +128,7 @@ opencli browser eval "fetch('...&fid=f237&po=1...').then(...)"
 
    ```bash
    # 取前三条的 keys，看哪些 key 的值在变（业务数据），哪些不变（常量/配置）
-   opencli browser eval "fetch('<url>').then(r=>r.json()).then(d=>d.data.diff.slice(0,3).map(x=>({f2:x.f2,f3:x.f3,f152:x.f152})))"
+   ppcli browser eval "fetch('<url>').then(r=>r.json()).then(d=>d.data.diff.slice(0,3).map(x=>({f2:x.f2,f3:x.f3,f152:x.f152})))"
    ```
 
 ---
@@ -154,7 +154,7 @@ opencli browser eval "fetch('...&fid=f237&po=1...').then(...)"
 
 1. **补进 `references/field-conventions.md`**：找到对应站点的表格加一行。下次直接查。
 2. **在 adapter 代码里留一条注释**：如果是实测推出来的不常见代号，写一行 `// f237 = convertible premium rate (verified 2026-04-20 against page)` 方便复核。
-3. **通过 `opencli browser verify` 验一次**：字段值能对上网页上眼见的数字。
+3. **通过 `ppcli browser verify` 验一次**：字段值能对上网页上眼见的数字。
 
 ---
 

package/skills/opencli-adapter-author/references/jsdom-fixture-pattern.md

@@ -40,9 +40,9 @@ the live extractor walks.
 
 1. **Capture** the page's HTML from a live verify run:
    ```bash
-   opencli browser open https://www.example.com/<page>
+   ppcli browser open https://www.example.com/<page>
    # In another shell, dump page.content():
-   opencli browser eval 'document.documentElement.outerHTML' \
+   ppcli browser eval 'document.documentElement.outerHTML' \
      > /tmp/raw-<command>.html
    ```
 

package/skills/opencli-adapter-author/references/site-memory.md

@@ -11,7 +11,7 @@ skills/opencli-adapter-author/references/site-memory/<site>.md
     — 公共种子。手写 + PR 审核进入。多 agent 共享的第一批起点。
     — 已铺：eastmoney / xueqiu / bilibili / tonghuashun
 
-~/.opencli/sites/<site>/
+~/.ppcli/sites/<site>/
     — 本地累积。agent 跑 adapter 过程里自动写入，跨 session 复用。
     — 不进 git，不进 PR。
 ```
@@ -52,16 +52,16 @@ skills/opencli-adapter-author/references/site-memory/<site>.md
 
 ---
 
-## Layer 2 — 本地工作目录（`~/.opencli/sites/<site>/`）
+## Layer 2 — 本地工作目录（`~/.ppcli/sites/<site>/`）
 
 agent 每跑一次相关 adapter 就可以自动写/读：
 
 ```
-~/.opencli/sites/<site>/
+~/.ppcli/sites/<site>/
   notes.md               — 累积笔记（时间戳 + 写入人 + 发现）
   endpoints.json         — 已验证的 endpoint 目录
   field-map.json         — 字段代号 → 含义（key 为字段代号，value 为 {meaning, verified_at, source}）
-  verify/                — `opencli browser verify` 期望值（值级校验锚点，每个 adapter 一份）
+  verify/                — `ppcli browser verify` 期望值（值级校验锚点，每个 adapter 一份）
     <cmd>.json
   fixtures/              — 完整响应样本（给字段对比 / 离线 replay；**调试时的原始 dump 也只能落在这里或 /tmp/**）
     <cmd>-<ts>.json
@@ -121,7 +121,7 @@ key = 字段代号（`f237` / `f152`），value 三件套：
 
 ### `verify/<cmd>.json` 格式（schema 锁死）
 
-每个 adapter 一份，`opencli browser verify <site>/<cmd>` 会自动读。**没有这份 = verify 只能证"能跑"，证不出数据对**——所以是必填产物。
+每个 adapter 一份，`ppcli browser verify <site>/<cmd>` 会自动读。**没有这份 = verify 只能证"能跑"，证不出数据对**——所以是必填产物。
 
 ```json
 {
@@ -194,16 +194,16 @@ key = 字段代号（`f237` / `f152`），value 三件套：
 ## runbook 里的读/写时机
 
 ```
-Step 2 开始前 → 读  ~/.opencli/sites/<site>/
+Step 2 开始前 → 读  ~/.ppcli/sites/<site>/
                 → 读  references/site-memory/<site>.md
                 命中后 → 不跳写 adapter，仍要跑 Step 5 (endpoint 验证) + Step 7 (字段抽查)
                         verified_at 超 30 天 → 当作过期，按冷启动走 Step 3 → 4
 
-Step 10 verify 首轮通过后 → 写 ~/.opencli/sites/<site>/verify/<cmd>.json
+Step 10 verify 首轮通过后 → 写 ~/.ppcli/sites/<site>/verify/<cmd>.json
                             - 先 `--write-fixture` 拿种子，再手改 patterns / notEmpty / rowCount
                             - 没这份后续 verify 挡不住数据错位，**必填**
 
-Step 11 肉眼对比通过后 → 写 ~/.opencli/sites/<site>/
+Step 11 肉眼对比通过后 → 写 ~/.ppcli/sites/<site>/
                         - endpoints.json：按 schema 追加或更新 verified_at
                         - field-map.json：只追加新 key，已有的不默默覆盖
                         - notes.md：顶部追加一段
@@ -214,7 +214,7 @@ Step 11 肉眼对比通过后 → 写 ~/.opencli/sites/<site>/
 
 ---
 
-## 不要写进 `~/.opencli/sites/` 的东西
+## 不要写进 `~/.ppcli/sites/` 的东西
 
 - 真实账户 cookie / token — 不要保存任何鉴权凭据
 - 用户私有数据（返回体里有个人敏感字段的 → 脱敏再存 fixtures）
@@ -222,10 +222,10 @@ Step 11 肉眼对比通过后 → 写 ~/.opencli/sites/<site>/
 
 ## 不要写进 **repo / adapter 目录** 的东西
 
-调试过程里的临时 dump（`.dbg-*.html` / `raw-*.json` / `sample-*` / `trace-*.txt`）**只能**落在 `~/.opencli/sites/<site>/fixtures/` 或系统 `/tmp/`。PR diff 会把 repo 根目录和 `clis/<site>/` 下的文件一起带走——别人 review 时看到一堆调试副产物会很烦。
+调试过程里的临时 dump（`.dbg-*.html` / `raw-*.json` / `sample-*` / `trace-*.txt`）**只能**落在 `~/.ppcli/sites/<site>/fixtures/` 或系统 `/tmp/`。PR diff 会把 repo 根目录和 `clis/<site>/` 下的文件一起带走——别人 review 时看到一堆调试副产物会很烦。
 
 ---
 
 ## 没有 site-memory 时
 
-新站点没对应 `.md`，也没本地目录 → 完整走 recon + discovery，跑完直接写 `~/.opencli/sites/<site>/`，后面就有了。
+新站点没对应 `.md`，也没本地目录 → 完整走 recon + discovery，跑完直接写 `~/.ppcli/sites/<site>/`，后面就有了。

package/skills/opencli-adapter-author/references/site-recon.md

@@ -9,7 +9,7 @@
 ## 一步诊断（推荐）
 
 ```bash
-opencli browser analyze <url>
+ppcli browser analyze <url>
 ```
 
 返回一份 JSON：
@@ -20,7 +20,7 @@ opencli browser analyze <url>
   "anti_bot": { "detected": false, "vendor": null, "evidence": [], "implication": "No known anti-bot signatures. Node-side fetch may work; try COOKIE first, fall back to browser-context fetch if blocked." },
   "initial_state": { "__INITIAL_STATE__": false, "__NUXT__": false, "__NEXT_DATA__": false, "__APOLLO_STATE__": false },
   "nearest_adapter": { "site": "xueqiu", "example_commands": ["xueqiu search", "xueqiu hot"], "reason": "2 existing adapters target this site — reuse strategy/cookie config" },
-  "recommended_next_step": "Pick the most specific JSON endpoint from `opencli browser network` and try a bare Node fetch with cookies; escalate to browser-context fetch only if blocked."
+  "recommended_next_step": "Pick the most specific JSON endpoint from `ppcli browser network` and try a bare Node fetch with cookies; escalate to browser-context fetch only if blocked."
 }
 ```
 
@@ -29,9 +29,9 @@ opencli browser analyze <url>
 ## 手动三步诊断（analyze 给不出明确结论时）
 
 ```bash
-opencli browser open <url>
-opencli browser wait time 2
-opencli browser network
+ppcli browser open <url>
+ppcli browser wait time 2
+ppcli browser network
 ```
 
 看 `network` 输出判：
@@ -46,7 +46,7 @@ opencli browser network
 
 分不清时参考下面五节的其他信号。
 
-**数据是 SPA / 异步加载时，`wait time 2` 可能不够**。改用 `opencli browser wait xhr '/api/path-fragment'` 直接等具体接口到场，比盲 `wait time 5` 更稳。
+**数据是 SPA / 异步加载时，`wait time 2` 可能不够**。改用 `ppcli browser wait xhr '/api/path-fragment'` 直接等具体接口到场，比盲 `wait time 5` 更稳。
 
 ---
 
@@ -62,7 +62,7 @@ opencli browser network
 **下一步**：`api-discovery.md` §1（network 精读）
 
 **注意 — Pattern A 命中不等于 strategy 选 `PAGE_FETCH`**：
-- 先看 `opencli browser analyze` 输出的 `api_candidates[]`：`verdict=likely_data` 的条目才是真候选；`verdict=noise`（analytics / beacon / personalization）不能算 API 信号
+- 先看 `ppcli browser analyze` 输出的 `api_candidates[]`：`verdict=likely_data` 的条目才是真候选；`verdict=noise`（analytics / beacon / personalization）不能算 API 信号
 - booking #1680 反例：17 个 JSON XHR 看起来像 Pattern A，但全是 analytics side-channel，最终走 `DOM_STATE` / `UI_SELECTOR`
 - replay 候选 endpoint 后，按 `strategy-selection.md` 的契约模型选 strategy；`PUBLIC_API` / `COOKIE_API` 都不通才考虑 `PAGE_FETCH`
 

package/skills/opencli-adapter-author/references/strategy-selection.md

@@ -1,8 +1,8 @@
 # Strategy Selection
 
-SKILL.md 顶层已给出 strategy gate 的 enum、表格和必填字段。本文件展开**为什么**这套 ladder 是按"契约"而不是"接口高度"组织的，以及具体怎么用 `opencli browser analyze` 的 `api_candidates` 证据填 strategy note。
+SKILL.md 顶层已给出 strategy gate 的 enum、表格和必填字段。本文件展开**为什么**这套 ladder 是按"契约"而不是"接口高度"组织的，以及具体怎么用 `ppcli browser analyze` 的 `api_candidates` 证据填 strategy note。
 
-进入条件：你已经按 `site-recon.md` 跑过 `opencli browser analyze`、按 `api-discovery.md` 抓过候选 endpoint。本文件是写 note 之前的最后一站。
+进入条件：你已经按 `site-recon.md` 跑过 `ppcli browser analyze`、按 `api-discovery.md` 抓过候选 endpoint。本文件是写 note 之前的最后一站。
 
 ---
 
@@ -50,7 +50,7 @@ SKILL.md 顶层已给出 strategy gate 的 enum、表格和必填字段。本文
 
 ## 3. 怎么把 `api_candidates` 转化为 strategy note 证据
 
-`opencli browser analyze <url>` 的输出里 `api_candidates[]` 字段，每条带：
+`ppcli browser analyze <url>` 的输出里 `api_candidates[]` 字段，每条带：
 
 ```json
 {

package/skills/opencli-adapter-author/references/success-rate-pitfalls.md

@@ -14,7 +14,7 @@
 
 **防御**：
 - `autofix` skill 现有纪律：**verify pattern 失败 = 收紧 adapter，不是收紧 fixture**（`opencli-autofix` SKILL.md §Rules for Patching 第 6 条）
-- 要改 fixture 的唯一合法理由：**站点本身换了格式**（例如 URL 规范迁移）。这种情况下在 `~/.opencli/sites/<site>/notes.md` 顶部写一段说明
+- 要改 fixture 的唯一合法理由：**站点本身换了格式**（例如 URL 规范迁移）。这种情况下在 `~/.ppcli/sites/<site>/notes.md` 顶部写一段说明
 
 ---
 
@@ -90,9 +90,9 @@
 **根因**：`wait time N` 是盲等。不同站点 JS 执行速度差很多。
 
 **防御**：
-- 数据是异步加载时**不用 `wait time`**，用 `opencli browser wait xhr '/api/path-fragment'`，等具体 XHR 到场再 `network`
+- 数据是异步加载时**不用 `wait time`**，用 `ppcli browser wait xhr '/api/path-fragment'`，等具体 XHR 到场再 `network`
 - 不确定 endpoint 路径时：先 `wait time 2 && network`，看到候选路径再转 `wait xhr` 确认
-- 首诊断用 `opencli browser analyze <url>` 一步拿 `json_responses` 数量——=0 时才真的是 Pattern C
+- 首诊断用 `ppcli browser analyze <url>` 一步拿 `json_responses` 数量——=0 时才真的是 Pattern C
 
 ---
 
@@ -113,7 +113,7 @@
 
 **现象**：本地开发时用自己的登录态验 endpoint 能通，PR 一合 verify fixture 跑在 CI 环境里立刻 401——顺手把样本数据也固化进了 fixture，看起来"一切正常"。
 
-**根因**：fixture 样本是带登录态跑出来的。存 `~/.opencli/sites/<site>/fixtures/*.json` 没脱敏，把 cookie / token / 自己的 uid / 昵称存了进去。
+**根因**：fixture 样本是带登录态跑出来的。存 `~/.ppcli/sites/<site>/fixtures/*.json` 没脱敏，把 cookie / token / 自己的 uid / 昵称存了进去。
 
 **防御**：
 - `site-memory.md` 的脱敏规则：存 fixtures 前去掉 cookie / token / 用户私有字段（手机号 / 邮箱 / 昵称 / uid）

package/skills/opencli-adapter-author/references/typed-errors.md

@@ -1,6 +1,6 @@
 # Typed Error Conventions
 
-OpenCLI 用 5 类 typed error 让 agent 能从 exit code 直接分辨"参数错 / 没数据 / 接口挂 / 要登录 / 超时"。silent `return []` / silent `return [{sentinel}]` / scalar sentinel (`'-'`) / `Math.max/min` silent clamp / `CliError('HTTP_ERROR')` 这些"绿但错"的写法都被 audit gate 抓得越来越紧（[`scripts/check-typed-error-lint.mjs`](../../../scripts/check-typed-error-lint.mjs) 的 baseline JSON 只能减不能加，新违例必须立刻收掉）。
+ppcli 用 5 类 typed error 让 agent 能从 exit code 直接分辨"参数错 / 没数据 / 接口挂 / 要登录 / 超时"。silent `return []` / silent `return [{sentinel}]` / scalar sentinel (`'-'`) / `Math.max/min` silent clamp / `CliError('HTTP_ERROR')` 这些"绿但错"的写法都被 audit gate 抓得越来越紧（[`scripts/check-typed-error-lint.mjs`](../../../scripts/check-typed-error-lint.mjs) 的 baseline JSON 只能减不能加，新违例必须立刻收掉）。
 
 每条 rule 都挂了真实 anti-pattern 反例（PR #1329 三轮迭代是主要素材库）。
 
@@ -209,7 +209,7 @@ if (!/id="postlist"/.test(html)) throw new EmptyResultError('1point3acres thread
 
 ## 6. Verify fixture 怎么挡这三类
 
-新写 fixture 时（`~/.opencli/sites/<site>/verify/<cmd>.json`）：
+新写 fixture 时（`~/.ppcli/sites/<site>/verify/<cmd>.json`）：
 
 - **rowCount.min ≥ 1**：保证 sentinel-row 不能伪装通过
 - **patterns**：核心 id 列加 `^\d+$` 类正则，挡 `tid: ''` / `pid: ''` 之类空字符串污染

package/skills/opencli-autofix/SKILL.md

@@ -1,10 +1,10 @@
 ---
 name: opencli-autofix
-description: Automatically fix broken OpenCLI adapters when commands fail. Load this skill when an opencli command fails — it guides you through collecting a trace artifact, patching the adapter, retrying, and filing an upstream GitHub issue after a verified fix. Works with any AI agent.
-allowed-tools: Bash(opencli:*), Bash(gh:*), Read, Edit, Write
+description: Automatically fix broken ppcli adapters when commands fail. Load this skill when an ppcli command fails — it guides you through collecting a trace artifact, patching the adapter, retrying, and filing an upstream GitHub issue after a verified fix. Works with any AI agent.
+allowed-tools: Bash(ppcli:*), Bash(gh:*), Read, Edit, Write
 ---
 
-# OpenCLI AutoFix — Automatic Adapter Self-Repair
+# ppcli AutoFix — Automatic Adapter Self-Repair
 
 When an `opencli` command fails because a website changed its DOM, API, or response schema, **automatically diagnose, fix the adapter, and retry** — don't just report the error.
 
@@ -13,11 +13,11 @@ When an `opencli` command fails because a website changed its DOM, API, or respo
 **Before starting any repair, check these hard stops:**
 
 - **`AUTH_REQUIRED`** (exit code 77) — **STOP.** Do not modify code. Tell the user to log into the site in Chrome.
-- **`BROWSER_CONNECT`** (exit code 69) — **STOP.** Do not modify code. Tell the user to run `opencli doctor`.
+- **`BROWSER_CONNECT`** (exit code 69) — **STOP.** Do not modify code. Tell the user to run `ppcli doctor`.
 - **CAPTCHA / rate limiting** — **STOP.** Not an adapter issue.
 
 **Scope constraint:**
-- **Only modify the file at `adapterSourcePath` in the trace `summary.md` front matter** — this is the authoritative adapter location (may be `clis/<site>/` in repo or `~/.opencli/clis/<site>/` for npm installs)
+- **Only modify the file at `adapterSourcePath` in the trace `summary.md` front matter** — this is the authoritative adapter location (may be `clis/<site>/` in repo or `~/.ppcli/clis/<site>/` for npm installs)
 - **Never modify** `src/`, `extension/`, `tests/`, `package.json`, or `tsconfig.json`
 
 **Retry budget:** Max **3 repair rounds** per failure. If 3 rounds of diagnose → fix → retry don't resolve it, stop and report what was tried.
@@ -25,12 +25,12 @@ When an `opencli` command fails because a website changed its DOM, API, or respo
 ## Prerequisites
 
 ```bash
-opencli doctor    # Verify extension + daemon connectivity
+ppcli doctor    # Verify extension + daemon connectivity
 ```
 
 ## When to Use This Skill
 
-Use when `opencli <site> <command>` fails with repairable errors:
+Use when `ppcli <site> <command>` fails with repairable errors:
 - **SELECTOR** — element not found (DOM changed)
 - **EMPTY_RESULT** — no data returned (API response changed)
 - **API_ERROR** / **NETWORK** — endpoint moved or broke
@@ -42,8 +42,8 @@ Use when `opencli <site> <command>` fails with repairable errors:
 
 `EMPTY_RESULT` — and sometimes a structurally-valid `SELECTOR` that returns nothing — is often **not an adapter bug**. Platforms actively degrade results under anti-scrape heuristics, and a "not found" response from the site doesn't mean the content is actually missing. Rule this out **before** committing to a repair round:
 
-- **Retry with an alternative query or entry point.** If `opencli xiaohongshu search "X"` returns 0 but `opencli xiaohongshu search "X 攻略"` returns 20, the adapter is fine — the platform was shaping results for the first query.
-- **Spot-check in a normal Chrome tab.** If the data is visible in the user's own browser but the adapter comes back empty, the issue is usually authentication state, rate limiting, or a soft block — not a code bug. The fix is `opencli doctor` / re-login, not editing source.
+- **Retry with an alternative query or entry point.** If `ppcli xiaohongshu search "X"` returns 0 but `ppcli xiaohongshu search "X 攻略"` returns 20, the adapter is fine — the platform was shaping results for the first query.
+- **Spot-check in a normal Chrome tab.** If the data is visible in the user's own browser but the adapter comes back empty, the issue is usually authentication state, rate limiting, or a soft block — not a code bug. The fix is `ppcli doctor` / re-login, not editing source.
 - **Look for soft 404s.** Sites like xiaohongshu / weibo / douyin return HTTP 200 with an empty payload instead of a real 404 when an item is hidden or deleted. The snapshot will look structurally correct. A retry 2-3 seconds later often distinguishes "temporarily hidden" from "actually gone".
 - **"0 results" from a search is an answer.** If the adapter successfully reached the search endpoint, got an HTTP 200, and the platform returned `results: []`, that is a valid answer — report it to the user as "no matches for this query" rather than patching the adapter.
 
@@ -54,7 +54,7 @@ Only proceed to Step 1 if the empty/selector-missing result is **reproducible ac
 Run the failing command with failure-retained trace enabled:
 
 ```bash
-opencli <site> <command> [args...] --trace retain-on-failure 2>trace-error.yaml
+ppcli <site> <command> [args...] --trace retain-on-failure 2>trace-error.yaml
 ```
 
 On failure, stderr contains the normal error envelope plus a small `trace` block:
@@ -68,9 +68,9 @@ trace:
   schemaVersion: 1
   opencliVersion: "..."
   traceId: "..."
-  dir: "/path/to/.opencli/profiles/default/traces/..."
-  summaryPath: "/path/to/.opencli/profiles/default/traces/.../summary.md"
-  receiptPath: "/path/to/.opencli/profiles/default/traces/.../receipt.json"
+  dir: "/path/to/.ppcli/profiles/default/traces/..."
+  summaryPath: "/path/to/.ppcli/profiles/default/traces/.../summary.md"
+  receiptPath: "/path/to/.ppcli/profiles/default/traces/.../receipt.json"
 ```
 
 Read `summaryPath` first. It is the LLM-oriented entry point and includes front matter:
@@ -126,13 +126,13 @@ Read the trace summary and the adapter source. Classify the root cause:
 
 ## Step 3: Explore the Current Website
 
-Use `opencli browser` to inspect the live website. **Never use the broken adapter** — it will just fail again.
+Use `ppcli browser` to inspect the live website. **Never use the broken adapter** — it will just fail again.
 
 ### DOM changed (SELECTOR errors)
 
 ```bash
 # Open the page and inspect current DOM
-opencli browser open https://example.com/target-page && opencli browser state
+ppcli browser open https://example.com/target-page && ppcli browser state
 
 # Look for elements that match the adapter's intent
 # Compare the snapshot with what the adapter expects
@@ -142,21 +142,21 @@ opencli browser open https://example.com/target-page && opencli browser state
 
 ```bash
 # Open page with network interceptor, then trigger the action manually
-opencli browser open https://example.com/target-page && opencli browser state
+ppcli browser open https://example.com/target-page && ppcli browser state
 
 # Interact to trigger API calls
-opencli browser click <N> && opencli browser network
+ppcli browser click <N> && ppcli browser network
 
 # Narrow to the request you care about by the fields its body should have
-opencli browser network --filter author,text,likes
+ppcli browser network --filter author,text,likes
 
 # Inspect specific API response (key is the `key` field from the default JSON output)
-opencli browser network --detail <key>
+ppcli browser network --detail <key>
 ```
 
 ## Step 4: Patch the Adapter
 
-Read the adapter source file at `adapterSourcePath` from the trace summary front matter and make targeted fixes. This path is authoritative — it may be in the repo (`clis/`) or user-local (`~/.opencli/clis/`).
+Read the adapter source file at `adapterSourcePath` from the trace summary front matter and make targeted fixes. This path is authoritative — it may be in the repo (`clis/`) or user-local (`~/.ppcli/clis/`).
 
 Use the `Read` tool on the exact path from summary.md front matter.
 
@@ -193,13 +193,13 @@ Use the `Read` tool on the exact path from summary.md front matter.
 3. **Prefer API over DOM scraping** — if you discover a JSON API during exploration, switch to it
 4. **Use `@jackwener/opencli/*` imports only** — never add third-party package imports
 5. **Test after patching** — run the command again to verify
-6. **Never relax `verify/<cmd>.json` fixtures to silence a failure.** A failing `patterns` / `notEmpty` / `mustNotContain` / `mustBeTruthy` rule means the adapter's output is broken. Tighten the adapter so it produces correct values; do not loosen the fixture to accept the broken values. The one legitimate reason to edit a fixture during repair is when the **site itself** changed shape (e.g. URL format migration) — in that case update the fixture and note the change in `~/.opencli/sites/<site>/notes.md`. Otherwise editing the fixture is covering up a silent correctness regression.
+6. **Never relax `verify/<cmd>.json` fixtures to silence a failure.** A failing `patterns` / `notEmpty` / `mustNotContain` / `mustBeTruthy` rule means the adapter's output is broken. Tighten the adapter so it produces correct values; do not loosen the fixture to accept the broken values. The one legitimate reason to edit a fixture during repair is when the **site itself** changed shape (e.g. URL format migration) — in that case update the fixture and note the change in `~/.ppcli/sites/<site>/notes.md`. Otherwise editing the fixture is covering up a silent correctness regression.
 
 ## Step 5: Verify the Fix
 
 ```bash
 # Run the command normally
-opencli <site> <command> [args...]
+ppcli <site> <command> [args...]
 ```
 
 If it still fails, go back to Step 1 and collect a fresh trace. You have a budget of **3 repair rounds** (trace → fix → retry). If the same error persists after a fix, try a different approach. After 3 rounds, stop and report what was tried.
@@ -223,12 +223,12 @@ If the retry **passes**, the local adapter has drifted from upstream. File a Git
 
 ```markdown
 ## Summary
-OpenCLI autofix repaired this adapter locally, and the retry passed.
+ppcli autofix repaired this adapter locally, and the retry passed.
 
 ## Adapter
 - Site: `<site>`
 - Command: `<command>`
-- OpenCLI version: `<version from opencli --version>`
+- ppcli version: `<version from ppcli --version>`
 
 ## Original failure
 - Error code: `<error_code>`
@@ -243,7 +243,7 @@ OpenCLI autofix repaired this adapter locally, and the retry passed.
 <1-2 sentence description of what you changed and why>
 ~~~
 
-_Issue filed by OpenCLI autofix after a verified local repair._
+_Issue filed by ppcli autofix after a verified local repair._
 ```
 
 2. **Ask the user before filing.** Show them the draft title and body. Only proceed if they confirm.
@@ -275,20 +275,20 @@ In all stop cases, clearly communicate the situation to the user rather than mak
 ## Example Repair Session
 
 ```
-1. User runs: opencli zhihu hot
+1. User runs: ppcli zhihu hot
    → Fails: SELECTOR "Could not find element: .HotList-item"
 
-2. AI runs: opencli zhihu hot --trace retain-on-failure 2>trace-error.yaml
+2. AI runs: ppcli zhihu hot --trace retain-on-failure 2>trace-error.yaml
    → Gets trace summary with final state and failed action evidence
 
 3. AI reads summary/state: page loaded but uses ".HotItem" instead of ".HotList-item"
 
-4. AI explores: opencli browser open https://www.zhihu.com/hot && opencli browser state
+4. AI explores: ppcli browser open https://www.zhihu.com/hot && ppcli browser state
    → Confirms new class name ".HotItem" with child ".HotItem-content"
 
 5. AI patches: Edit adapter at `adapterSourcePath` — replace ".HotList-item" with ".HotItem"
 
-6. AI verifies: opencli zhihu hot
+6. AI verifies: ppcli zhihu hot
    → Success: returns hot topics
 
 7. AI prepares upstream issue draft, shows it to the user