kimi-webbridge-dsh 1.4.8 → 1.4.10

package/README.md

@@ -1,118 +1,61 @@
 # kimi-webbridge-dsh
 
-Kimi 浏览器扩展（**Kimi Browser Extension**，原 Kimi WebBridge）的 dsh 插件（**基础设施 + 技能注册**）。插件本身不代理浏览器操作，而是负责：
+把 [Kimi 浏览器扩展](https://www.kimi.com/products/kimi-webbridge)（Kimi Browser Extension，原 Kimi WebBridge）接进 dsh：**供应守护进程 + 注册官方技能包**。插件本身不代理浏览器操作——浏览器能力全部由官方技能文档描述，模型照它驱动本地守护进程。
 
-1. **保障桥完好**：守护进程二进制缺失时自动从官方 CDN 下载（跨平台、SHA-256 校验），守护进程未运行时自动拉起；
-2. **注册 skill**：启动时把随包内置的 `skill.md` 注册为运行时 skill（`ctx.skills.register`，`source: bundled`），模型通过 `skill` 工具加载它，按文档用 HTTP 行使全部浏览器能力（25 个动作）；
-3. **注册一个瘦工具 `kimi_webbridge`**：只有 `status` 一个动作（守护进程 + 扩展健康检查），作为其他 skill（如 x-cli 系列）的桥健康锚点。
-
-## 架构
-
-| 层 | 角色 |
+| 职责 | 说明 |
 |---|---|
-| 插件（本包） | 守护进程二进制供应 + 自动启动；apply() 时注册运行时 skill |
-| skill（运行时注册，随包内置 `skill.md`） | 模型经 `skill` 工具加载，按文档驱动守护进程：navigate / find_tab / snapshot / read_page / find / click / mouse_click / hover / drag / fill / key_type / send_keys / select_option / dialog / evaluate / cdp / screenshot / network / upload / save_as_pdf / scroll / wait / list_tabs / close_tab / close_session |
-| 瘦工具 `kimi_webbridge` | 仅 `status` 健康检查（自带守护进程自愈） |
+| 守护进程供应 | 按官方清单下载对应平台二进制（SHA-256 校验），未运行时自动拉起；落在 `~/.kimi-webbridge/bin/` |
+| 技能注册 | 随包内置官方技能文档（含 `references/`），启动时注册为 `kimi-webbridge` skill |
+| 健康锚点 | 瘦工具 `kimi_webbridge`，只有 `status` 一个动作（守护进程 + 扩展 + 已装技能副本） |
 
-## 安装（一次性，零手工配置）
+## 安装
 
-```powershell
+```bash
 dsh plugin --profile web add kimi-webbridge-dsh
 ```
 
-本包声明了 `dsh.bundle.patch`：`dsh plugin add` 会自动把它登记进 profile 的 bundle 层（`dsh.profile.bundles`），**无需手动编辑任何 cordis.patch.yml**。重启一次 `dsh web`：启动时插件把 skill 注册进运行时 registry，新会话的 skill 目录中即出现 `kimi-webbridge`。
-
-## 用法（模型视角）
-
-1. 需要真实浏览器时，加载 `kimi-webbridge` skill（触发词：browser / webpage / open URL / 浏览器 / 网页 / 截图……）；
-2. 先用 `kimi_webbridge` 工具的 `status` 确认守护进程 + 扩展健康（插件自动补装/拉起二进制）；
-3. 按 skill 文档向 `POST http://127.0.0.1:10086/command` 发送 `{action, args, session}` 行使浏览器能力。Windows 上必须用临时文件承载 JSON（shell 会破坏非 ASCII 文本），skill 里有完整说明。
-
-## 依赖
-
-- **浏览器扩展**：需安装并连接 Kimi 浏览器扩展（Kimi Browser Extension，原 Kimi WebBridge）—— 从 [Chrome Web Store](https://chromewebstore.google.com/detail/kimi-webbridge/fldmhceldgbpfpkbgopacenieobmligc) 安装（`kimi_webbridge status` 显示 `extension_connected: true`；扩展版本应与守护进程一致）。
-- **守护进程**：无需预装。插件在守护进程不可达时会自动启动它；二进制缺失时会自动从官方 CDN（cdn.kimi.com，与官方安装器同源）下载安装——全程无需手动运行任何安装命令。
-
-### 版本对齐
-
-插件按官方 CDN 的 `version.json` 供应守护进程，因此始终与官方发布对齐；上表的存档哈希对应 **v2.0.8**（守护进程与扩展的产品名已更名为 Kimi Browser Extension）。
-
-`kimi_webbridge` 的 `status` 会原样透出守护进程的 `update_available`（有更新时）与 `version_mismatch`（守护进程与扩展不同步时）字段。按官方约定：这两个字段出现时**不要**自动执行 `upgrade` / `stop` / `restart` / `uninstall`（它们会杀掉正在运行的守护进程），而应把字段里的 `command` 原样交给用户执行。
-
-### 兼容性（dsh 版本）
-
-插件只 import 宿主提供的 `@deepseek-ai/*` 包，并声明为 `peerDependencies`（不打包自己的副本），所以与 dsh 版本解耦；已在 **dsh 0.1.5-rc.1** 上验证（工具注册、skill 注册、system prompt 注入、守护进程自愈均正常）。
-
-| peer | 声明范围 |
-|---|---|
-| `@deepseek-ai/dsh-tools` | `^0.1.0-rc.7 \|\| ^0.1.1-rc.2 \|\| ^0.1.2-rc.1 \|\| ^0.1.5-rc.1` |
-| `@deepseek-ai/dsh-llm` | 同上 |
-| `@deepseek-ai/schemastery` | `^3.18.1` |
-
-⚠️ **预发布版必须逐个元组写进范围。** node-semver 只在范围里存在一个**与该版本 `major.minor.patch` 完全一致、且自身带预发布标签**的比较符时才放行预发布版；因此看起来很宽的 `>=0.0.1-rc.1 <0.2.0` 会静默排除所有预发布构建，用户会遇到 `ERESOLVE`。dsh 发布新的预发布系列（如 `0.1.6-rc.1`）时，需要同步追加一个 `|| ^<新元组>-rc.N` 分支。
-
-### 跨平台支持
+自带 bundle patch，无需手改 `cordis.patch.yml`；重启一次 dsh 生效。
 
-插件根据运行时硬件自动选择并下载对应的守护进程二进制（含 SHA-256 校验，与官方清单核对后才安装）：
-
-| 平台 | 二进制 | 存档哈希（v2.0.8） |
-|---|---|---|
-| Windows x64 | `windows-amd64.exe` | `eec1976d…433c6` |
-| macOS arm64 | `darwin-arm64` | `04532d77…0a261` |
-| macOS x64 | `darwin-amd64` | `931769e9…8cc0c` |
-| Linux arm64 | `linux-arm64` | `ee201b44…34a3c` |
-| Linux x64 | `linux-amd64` | `61de010c…ce271` |
+## 用法
 
-不支持的平台组合会明确报错并提示手动安装。
+1. 需要真实浏览器时加载 `kimi-webbridge` skill（触发词：browser / 网页 / 截图 / ……，也包括"给某个网站做个可复用 CLI"）；
+2. 先调 `kimi_webbridge` 的 `status` 确认守护进程 + 扩展健康（会自动补装 / 拉起）；
+3. 按 skill 文档 `POST http://127.0.0.1:10086/command`，请求体 `{action, args, session}`；
+4. Windows 必须用临时文件承载 JSON 请求体（shell 会破坏非 ASCII），skill 里有说明。
 
-### 守护进程二进制位置
-
-下载到用户目录（与官方安装器路径一致，无需管理员权限；按用户隔离）：
-
-| 平台 | 最终路径 |
-|---|---|
-| Windows | `%USERPROFILE%\.kimi-webbridge\bin\kimi-webbridge.exe` |
-| macOS / Linux | `~/.kimi-webbridge/bin/kimi-webbridge`（自动 `chmod 755`） |
+工具清单、参数与注意事项一律以 skill 文档为准。
 
-下载流程：先落盘为临时文件（`kimi-webbridge(.exe).tmp-{pid}-{时间戳}`）→ SHA-256 与官方清单核对 → 匹配才重命名为最终路径；不匹配则删除临时文件并中止，任何未验证的二进制都不会留在磁盘。该路径与官方安装器 / `kimi-webbridge upgrade` 命令完全一致，插件、官方工具共用同一份守护进程。
+## 配置
 
-## 功能（skill 承载，25 个动作）
-
-| 能力 | 动作 |
-|---|---|
-| 导航 / 标签 | `navigate`、`find_tab`、`list_tabs`、`close_tab`、`close_session` |
-| 读页面 | `snapshot`（accessibility tree + `@e` 引用）、`read_page`（全文文本）、`find`（页面内搜索） |
-| 点击 / 输入 | `click`、`mouse_click`（真实鼠标，可信输入）、`hover`、`drag`、`fill`、`key_type`（键入，可信）、`send_keys`（组合键）、`select_option`、`dialog` |
-| JS / CDP | `evaluate`（页面上下文）、`cdp`（chrome.debugger 直通） |
-| 截图 / PDF | `screenshot`、`save_as_pdf`（写文件、返回路径） |
-| 网络 / 上传 | `network`（请求捕获）、`upload`（文件上传） |
-| 滚动 / 等待 | `scroll`（滚动/定位）、`wait`（等待文本/选择器条件） |
-| 会话分组 | 一个任务 = 一个 session = 一个标签页组，可一键清理 |
-
-## 配置（可选）
-
-在 profile 的用户补丁层（`~/.dsh/profiles/web/cordis.patch.yml`）按 id 覆盖配置（替换整份 config，未给出的键取默认值）：
+在 `~/.dsh/profiles/web/cordis.patch.yml` 按 id 覆盖（整份替换，未给出的键取默认）：
 
 ```yaml
 - id: kimi-webbridge
   config:
-    autoInstallDaemon: false
-    daemonVersion: v2.0.8
+    daemonVersion: v2.0.22
 ```
 
-| 配置项 | 默认值 | 说明 |
+| 键 | 默认 | 说明 |
 |---|---|---|
 | `baseUrl` | `http://127.0.0.1:10086` | 守护进程地址 |
-| `requestTimeoutMs` | `120000` | 单次请求超时（毫秒） |
-| `autoStartDaemon` | `true` | 守护进程未运行时自动启动 |
-| `autoInstallDaemon` | `true` | 二进制缺失时自动从官方 CDN 下载安装（关闭后需手动[从官方安装](https://www.kimi.com/products/kimi-webbridge)） |
-| `daemonVersion` | `latest` | 自动安装时的版本（可固定如 `v2.0.8`，需带 `v` 前缀与 CDN 路径一致；建议与扩展版本匹配） |
-| `registerSkill` | `true` | 启动时把 `kimi-webbridge` skill 注册进运行时 registry（关闭后不注册） |
+| `requestTimeoutMs` | `120000` | 单次请求超时（ms） |
+| `autoStartDaemon` | `true` | 未运行时自动启动 |
+| `autoInstallDaemon` | `true` | 二进制缺失时自动下载安装 |
+| `daemonVersion` | `latest` | 自动安装的版本，带 `v` 前缀，建议与扩展版本一致 |
+| `registerSkill` | `true` | 注册 `kimi-webbridge` skill |
+
+## 要求与提示
+
+- **浏览器扩展**：需自行安装并连接（[Chrome Web Store](https://chromewebstore.google.com/detail/kimi-webbridge/fldmhceldgbpfpkbgopacenieobmligc)）；`status` 显示 `extension_connected: true` 即可用。
+- **守护进程**：无需预装，插件自动供应；与官方安装器共用 `~/.kimi-webbridge/bin/`（Windows `%USERPROFILE%\.kimi-webbridge\bin\`）。
+- **dsh 版本**：只 import 宿主提供的 `@deepseek-ai/*`（声明为 peer，不打包副本），已在 **0.2.0-rc.1** 验证，peer 范围覆盖 0.1.0-rc.7 起的整个 0.1.x / 0.2.x。dsh 换代后如果 `kimi_webbridge` 工具、系统提示段落和 skill 一起消失，通常就是插件版本还没覆盖新 dsh——升级本插件再重启即可。
+- **`status` 里的 `version_mismatch` / `update_available` / `skill_mismatch`**：把其中的 `command` 原样交给用户执行；**不要**自行 `upgrade` / `stop` / `restart` / `uninstall`，它们会杀掉正在运行的守护进程。
+- `status.skills` 为空是正常的：守护进程只识别 Claude Code / Codex / Hermes / OpenClaw / Kimi，dsh 拿到技能的唯一途径就是本插件。
 
 ## 移除
 
-```powershell
+```bash
 dsh plugin --profile web remove kimi-webbridge-dsh
 ```
 
-自动从 bundle 层移除（无需手动删补丁行），重启后生效。skill 为运行时注册，移除插件即随之消失，无残留文件。
+重启后生效；skill 随插件注册，移除即消失。

package/lib/index.js

@@ -17,8 +17,9 @@ import { fileURLToPath } from "node:url";
  * Infrastructure-only bridge plugin:
  * - provisions the daemon binary (official Kimi CDN, SHA-256 verified) and
  *   auto-starts it, keeping the bridge healthy;
- * - registers the packaged skill.md (the full 25-action HTTP protocol for
- *   driving the user's real browser) as a runtime skill via ctx.skills;
+ * - registers the skill package vendored under skill/ (the official upstream
+ *   SKILL.md plus its references/) as a runtime skill via ctx.skills, with the
+ *   directory exposed as its resource base;
  * - registers one thin tool `kimi_webbridge` whose only action is `status`
  *   (daemon + extension health check) as an anchor for other skills.
  *
@@ -94,15 +95,149 @@ function manualInstallHint() {
 /** Skill identity registered into the host skill registry. */
 const SKILL_NAME = "kimi-webbridge";
 
-/** Model-facing routing description shown in the skill catalog. */
-const SKILL_DESCRIPTION = `Drive the user's real browser (with their login sessions) through the local kimi-webbridge daemon at 127.0.0.1:10086: navigate, click, type, read, screenshot, evaluate JS, raw CDP, network capture, file upload, PDF export, and tab/session management. Use this skill whenever the user wants to interact with websites, automate browser tasks, scrape web content, or do anything that needs a real browser — even simple-sounding requests. Also use when the user mentions "browser", "webpage", "open URL", "screenshot", or asks to read/interact with any website (浏览器 / 网页 / 打开网页 / 截图 / 页面).`;
+/** Vendored upstream skill package: `skill/SKILL.md` plus its `references/` tree. */
+const SKILL_DIR = fileURLToPath(new URL("../skill/", import.meta.url));
+
+/** The upstream instruction file inside {@link SKILL_DIR}. */
+const SKILL_PATH = join(SKILL_DIR, "SKILL.md");
+
+/**
+ * Extra routing triggers appended to the description upstream ships.
+ *
+ * The official description is English; the Chinese terms are what the same users type,
+ * and the CLI trigger is one upstream only states at the very end of a long paragraph.
+ * Keeping them here (rather than rewriting the vendored frontmatter) is what lets
+ * `skill/UPSTREAM.json` stay byte-exact against the CDN.
+ */
+const SKILL_EXTRA_TRIGGERS = "Also use to build a reusable CLI or script that automates a specific website. Also use when the user mentions 浏览器 / 网页 / 打开网页 / 截图 / 页面.";
+
+/**
+ * Length of the routing line dsh renders in its model-facing skill catalog.
+ *
+ * `dsh-tool-skill` builds that line from the description and truncates it at
+ * `catalogDescriptionMaxLength` (500 by default), silently dropping the rest — so the
+ * registered description has to fit, or the model never sees the triggers above.
+ */
+const CATALOG_DESCRIPTION_LIMIT = 500;
+
+/**
+ * Used only when the vendored SKILL.md frontmatter cannot be parsed; a missing description
+ * must never disable the skill. Carries no triggers of its own — {@link composeSkillDescription}
+ * appends the shared ones either way.
+ */
+const FALLBACK_SKILL_DESCRIPTION = "Drive the user's real browser through the local kimi-webbridge daemon (127.0.0.1:10086) using their login sessions: navigate, read pages and iframes, click, type, screenshot, evaluate JS, raw CDP, network capture, uploads, PDF export, tabs and sessions. Use whenever a task needs a real browser.";
+
+/**
+ * The section appended after the vendored upstream instructions.
+ *
+ * Everything above it is upstream's document verbatim; this carries only the deltas
+ * that are true in dsh (tool names, the status anchor, where references live), plus
+ * the upstream release the copy came from — which stripping the frontmatter would
+ * otherwise hide from the model. Keeping the split means refreshing the skill never
+ * has to merge prose.
+ *
+ * @param version - the upstream version stamped in the vendored SKILL.md, when known.
+ * @returns the markdown block to append.
+ */
+function dshRuntimeSection(version) {
+  const provenance = typeof version === "string" && version.length > 0
+    ? `This skill is vendored verbatim from the official upstream release v${version}; \`skill/UPSTREAM.json\` records its digests.`
+    : "This skill is vendored verbatim from the official upstream release.";
+  return `
+
+## dsh runtime notes
+
+The kimi-webbridge-dsh plugin adds these notes for this harness; where they conflict with the text above, they win.
+
+${provenance}
+
+- **Health check first**: before driving the browser, call the \`kimi_webbridge\` tool with action \`status\`. It reports daemon + extension health and auto-provisions this machine — it installs the daemon binary from the official CDN (SHA-256 checked against the release manifest) when it is missing, starts it, and retries once. The binary lives at \`~/.kimi-webbridge/bin/kimi-webbridge\` and is not on \`PATH\`; running it with \`start\` yourself is also safe and idempotent.
+- **Tool names in this harness**: file tools are \`read\`, \`read_image\`, \`write\` and \`edit\`; the shell tool is \`bash\`; \`kimi_webbridge\` is the status anchor. Where the text above says "open it with the Read tool" for a screenshot or PDF, use \`read_image\`.
+- **References**: paths such as \`references/operations.md\` and \`references/cli-creator/workflow.md\` resolve against the base directory reported when this skill loads. Read them with the \`read\` tool.
+- **Skills installer**: the daemon's own \`install-skill\` / \`upgrade\` skill refresh targets the runtimes it detects (Claude Code, Codex, Hermes, OpenClaw, Kimi) — dsh is not one of them. This plugin is how dsh receives this skill, so never run \`install-skill\` on dsh's behalf.
+- **Notices**: when \`status\` reports \`version_mismatch\`, \`update_available\` or \`skill_mismatch\`, hand the user that field's \`command\` verbatim. Never run \`stop\` / \`restart\` / \`upgrade\` / \`uninstall\` yourself — they kill the user's running daemon.
+`;
+}
+
+/**
+ * Parse the two frontmatter fields this plugin needs from the vendored SKILL.md.
+ *
+ * A deliberately small YAML subset parser: the upstream frontmatter is a fixed shape
+ * (`name`, a `description` block scalar, `metadata.version`), and pulling in a YAML
+ * dependency for three fields would cost more than it buys.
+ *
+ * @param raw - the SKILL.md text.
+ * @returns the description, stamped version and body without frontmatter; fields are null when absent.
+ */
+function parseFrontmatter(raw) {
+  const match = /^---\r?\n([\s\S]*?)\r?\n---(?:\r?\n|$)/u.exec(raw);
+  if (match === null) return { description: null, version: null, body: raw.trim() };
+  let description = null;
+  let version = null;
+  let block = null;
+  for (const line of match[1].split(/\r?\n/u)) {
+    if (block !== null) {
+      if (/^\s+\S/u.test(line)) {
+        block.push(line.trim());
+        continue;
+      }
+      description = block.join(" ").trim();
+      block = null;
+    }
+    if (/^description:\s*[|>]\s*$/u.test(line)) {
+      block = [];
+      continue;
+    }
+    const inline = /^description:\s*(\S.*)$/u.exec(line);
+    if (inline !== null) {
+      description = inline[1].trim();
+      continue;
+    }
+    const stamped = /^\s+version:\s*"?([^"\s]+)"?\s*$/u.exec(line);
+    if (stamped !== null) version = stamped[1];
+  }
+  if (block !== null) description = block.join(" ").trim();
+  return { description, version, body: raw.slice(match[0].length).trim() };
+}
+
+/**
+ * Compose the registered routing description inside dsh's catalog budget.
+ *
+ * Upstream's own words lead — they name the product and the core capability — while the
+ * added triggers are reserved their space up front, so the truncation dsh applies can
+ * never eat them. When upstream does not fit, it is cut at a sentence boundary rather
+ * than mid-word, which is what dsh's own truncation would do to the whole line.
+ *
+ * @param upstream - the description from the vendored SKILL.md frontmatter.
+ * @returns a description of at most {@link CATALOG_DESCRIPTION_LIMIT} characters.
+ */
+function composeSkillDescription(upstream) {
+  const triggers = SKILL_EXTRA_TRIGGERS;
+  const text = upstream.replace(/\s+/gu, " ").trim();
+  const room = CATALOG_DESCRIPTION_LIMIT - triggers.length - 1;
+  if (text.length <= room) return `${text} ${triggers}`;
+  const visible = text.slice(0, room + 1);
+  const sentenceEnd = Math.max(visible.lastIndexOf(". "), visible.lastIndexOf("。"));
+  if (sentenceEnd > 0) return `${visible.slice(0, sentenceEnd + 1)} ${triggers}`;
+  return `${visible.slice(0, room).replace(/[\s,;:.]+$/u, "")} ${triggers}`;
+}
 
 /**
- * Read the packaged skill body (markdown instructions, no frontmatter).
- * @returns the skill body text.
+ * Load the vendored upstream skill package.
+ *
+ * @returns the routing description and the content to register (upstream body plus the
+ * dsh runtime notes, which carry the upstream release the copy came from).
  */
-function readSkillBody() {
-  return readFileSync(join(dirname(fileURLToPath(import.meta.url)), "..", "skill.md"), "utf8");
+function readSkillPackage() {
+  const { description, version, body } = parseFrontmatter(readFileSync(SKILL_PATH, "utf8"));
+  const fallback = typeof description !== "string" || description.length === 0;
+  if (fallback) {
+    console.warn(`kimi-webbridge: ${SKILL_PATH} has no frontmatter description; falling back to the built-in routing text`);
+  }
+  return {
+    description: composeSkillDescription(fallback ? FALLBACK_SKILL_DESCRIPTION : description),
+    content: `${body}${dshRuntimeSection(version)}`
+  };
 }
 
 /** Underlying transport `code` for a rejected `fetch`, which buries it one level down in `cause`. */
@@ -360,11 +495,15 @@ function apply(ctx, config = {}) {
 
   if (registerSkill) {
     try {
+      const skill = readSkillPackage();
+      // resourceBase is what lets the model resolve the upstream document's
+      // "references/..." links: the host renders it as this skill's base directory.
       ctx.skills.register({
         name: SKILL_NAME,
-        description: SKILL_DESCRIPTION,
+        description: skill.description,
         source: "bundled",
-        content: readSkillBody()
+        content: skill.content,
+        resourceBase: { kind: "directory", path: SKILL_DIR }
       });
     } catch (error) {
       // Never break plugin startup over skill registration; the status tool still works.
@@ -375,12 +514,12 @@ function apply(ctx, config = {}) {
   ctx.systemPrompt.section({
     name: "tool:kimi-webbridge",
     order: 105,
-    text: "kimi-webbridge: the local kimi-webbridge daemon (127.0.0.1:10086) bridges to the user's real browser with their login sessions; this plugin provisions the daemon binary and keeps it running. The `kimi_webbridge` tool is only the status health check (daemon + extension). For any browser automation (navigation, reading pages, clicking, typing, filling forms, screenshots, JS evaluation, CDP, uploads, PDF export, tab management), load the kimi-webbridge skill and drive the daemon over HTTP exactly as that skill documents."
+    text: "kimi-webbridge: the local kimi-webbridge daemon (127.0.0.1:10086) bridges to the user's real browser with their login sessions; this plugin provisions the daemon binary, keeps it running, and registers the official upstream skill package (with its references/ directory). The `kimi_webbridge` tool is only the status health check (daemon + extension + installed skill copies). For any browser work — navigation, reading pages and iframes, clicking, typing, form filling, trusted input via CDP focus emulation, screenshots, JS/CDP, network capture, uploads, PDF export, tabs and sessions, and building a reusable CLI for a website — load the kimi-webbridge skill and drive the daemon over HTTP exactly as that skill (and its references) documents."
   });
 
   ctx.tools.register(defineTool({
     name: "kimi_webbridge",
-    description: "Health check for the local kimi-webbridge bridge (daemon at " + baseUrl + " and the browser extension). The plugin auto-provisions the daemon: it installs the binary from the official Kimi CDN when missing and starts it, then retries once. Returns daemon version/uptime, extension_connected, plus version (daemon) and extension_version (extension) — a mismatch means the components have drifted (the extension likely auto-updated). This tool does NOT perform browser actions — for navigation, clicking, typing, reading pages, screenshots, JS evaluation, CDP, uploads, PDF export and tab management, load the kimi-webbridge skill, which documents the full 25-action HTTP protocol (POST " + baseUrl + "/command) to drive the user's real browser.",
+    description: "Health check for the local kimi-webbridge bridge (daemon at " + baseUrl + " and the browser extension). The plugin auto-provisions the daemon: it installs the binary from the official Kimi CDN when missing and starts it, then retries once. Returns daemon version/uptime, extension_connected, plus version (daemon) and extension_version (extension) — a mismatch means the components have drifted (the extension likely auto-updated) — and the upstream notices skills, version_mismatch, update_available and skill_mismatch. This tool does NOT perform browser actions — for navigation, reading pages and iframes, clicking, typing, screenshots, JS/CDP, trusted input, network capture, uploads, PDF export, tabs/sessions, or building a reusable CLI for a website, load the kimi-webbridge skill, which documents the daemon's tool protocol (POST " + baseUrl + "/command) and ships its references/ alongside.",
     parameters: {
       action: {
         type: "string",
@@ -405,7 +544,7 @@ function apply(ctx, config = {}) {
             type: "text",
             text: daemonAhead
               ? `⚠️ the browser extension (${value.extension_version}) is behind the daemon (${value.version}) — ask the user to update the extension, then retry.`
-              : `⚠️ version drift: daemon ${value.version} vs extension ${value.extension_version} — the browser extension likely auto-updated ahead of the daemon. Re-audit the 25-action skill list and update the daemon.`
+              : `⚠️ version drift: daemon ${value.version} vs extension ${value.extension_version} — the browser extension likely auto-updated ahead of the daemon. Re-audit the skill against the new tool surface and update the daemon.`
           });
         }
         // The official convention: surface the release fields verbatim and let the
@@ -417,6 +556,22 @@ function apply(ctx, config = {}) {
             text: `⬆️ update available: daemon ${update.current ?? "?"} → ${update.latest ?? "?"} — tell the user and hand them this command verbatim; do NOT run it yourself: ${update.command}`
           });
         }
+        // Installed skill copies the daemon found in the agent runtimes it manages.
+        // dsh is not one of them, so an empty list is the normal dsh result.
+        const skills = value?.skills;
+        if (Array.isArray(skills) && skills.length > 0) {
+          const listed = skills
+            .map((entry) => `${entry?.agent ?? "?"}@${entry?.version ?? "?"}${typeof entry?.path === "string" ? ` (${entry.path})` : ""}`)
+            .join("; ");
+          blocks.push({ type: "text", text: `installed skill copies: ${listed}` });
+        }
+        const skillMismatch = value?.skill_mismatch;
+        if (skillMismatch !== null && typeof skillMismatch === "object" && typeof skillMismatch.command === "string") {
+          blocks.push({
+            type: "text",
+            text: `⚠️ installed kimi-webbridge skill copies are out of step with the extension (${skillMismatch.extension_version ?? "?"}) — tell the user and hand them this command verbatim; do NOT run it yourself: ${skillMismatch.command}`
+          });
+        }
         return blocks;
       }
     },

package/package.json

@@ -1,9 +1,12 @@
 {
   "name": "kimi-webbridge-dsh",
-  "description": "Infrastructure plugin for dsh: provisions the kimi-webbridge daemon (auto-install with SHA-256 verification, auto-start) and registers the kimi-webbridge skill that documents the full 25-action browser protocol. Registers a thin kimi_webbridge tool (status health check only).",
-  "version": "1.4.8",
+  "description": "Infrastructure plugin for dsh: provisions the kimi-webbridge daemon (auto-install with SHA-256 verification, auto-start) and registers the official upstream skill package (SKILL.md + references/, vendored from the CDN and digest-pinned) as a runtime skill. Registers a thin kimi_webbridge tool (status health check only).",
+  "version": "1.4.10",
   "license": "MIT",
-  "repository": { "type": "git", "url": "git+https://github.com/cfanmaoli/kimi-webbridge-dsh.git" },
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/cfanmaoli/kimi-webbridge-dsh.git"
+  },
   "type": "module",
   "keywords": [
     "dsh",
@@ -12,6 +15,7 @@
     "webbridge",
     "browser",
     "automation",
+    "skill",
     "plugin",
     "cordis"
   ],
@@ -32,13 +36,18 @@
   },
   "files": [
     "lib/index.js",
-    "skill.md",
+    "skill",
+    "scripts",
     "README.md",
     "cordis.patch.yml"
   ],
+  "scripts": {
+    "check": "node --check lib/index.js && node scripts/fetch-upstream-skill.mjs --verify-local",
+    "fetch-skill": "node scripts/fetch-upstream-skill.mjs"
+  },
   "peerDependencies": {
-    "@deepseek-ai/dsh-tools": "^0.1.0-rc.7 || ^0.1.1-rc.2 || ^0.1.2-rc.1 || ^0.1.5-rc.1",
-    "@deepseek-ai/dsh-llm": "^0.1.0-rc.7 || ^0.1.1-rc.2 || ^0.1.2-rc.1 || ^0.1.5-rc.1",
+    "@deepseek-ai/dsh-tools": "^0.1.0-rc.7 || ^0.1.1-rc.2 || ^0.1.2-rc.1 || ^0.1.5-rc.1 || ^0.1.7-rc.2 || ^0.2.0-rc.1",
+    "@deepseek-ai/dsh-llm": "^0.1.0-rc.7 || ^0.1.1-rc.2 || ^0.1.2-rc.1 || ^0.1.5-rc.1 || ^0.1.7-rc.2 || ^0.2.0-rc.1",
     "@deepseek-ai/schemastery": "^3.18.1"
   }
-}
+}