Merge remote-tracking branch 'origin/master' into codex/dsh-home-path

# Conflicts:
#	packages/ui/app-boot/README.i18n.yaml
#	packages/ui/app-boot/README.md
#	packages/ui/app-boot/README.zh.md
#	packages/ui/app-boot/src/index.ts

apps/cli/README.i18n.yaml

@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write apps/cli/README.md
-README.md: fc5bce195fb10872c605cada1bcb3ed79380265c
-README.zh.md: 2fb1272231abb02e145e5c9925362de27307ca86
+README.md: cf038ad19c631721c7b3182ffe83e75e3837d9ba
+README.zh.md: c790f973a9ab0071253ab161bd8bc7835ebb2e02

apps/cli/README.md

@@ -3,7 +3,7 @@
 English | [中文](README.zh.md)
 
 
-Argv is parsed once through a [Commander](https://github.com/tj/commander.js) adapter ([`src/args.ts`](src/args.ts)): one program whose default (no subcommand) is the TUI/headless surface (`--config`, `-p`/`--prompt`, `--resume`), whose `meta` subcommand is the same TUI over this checkout, whose `upgrade` subcommand is an option-less guided-session entry, and whose `web` subcommand is the browser UI. `src/bin.ts` switches on the resolved mode and dynamic-imports only that mode's module. `dsh --help` lists every mode and `dsh web --help` renders the web usage, `dsh --version` prints this app's version, and an unknown option or a mistyped `--resume` fails loud (stderr, exit 1) instead of misrouting. Every subcommand that shares no option with the default surface — `upgrade`, `web`, `meta` — rejects a leaked `--config`/`-p`/`--resume` rather than running and dropping it. `dsh web`'s `--host`/`--port` are unvalidated pass-through overrides: the `dsh-host-webserver` schema is the single source of both the default (the shipped Web overlay value when a flag is absent) and validity, and rejects a bad value at boot. `--trusted-host` appends named authorities for the /api browser-trust fence; an all-interfaces bind additionally derives the machine's LAN IP literals itself ([`src/app-cli-entry.ts`](src/app-cli-entry.ts)), so the printed LAN URL works without flags.
+Argv is parsed once through a [Commander](https://github.com/tj/commander.js) adapter ([`src/args.ts`](src/args.ts)): one program whose default (no subcommand) is the TUI/headless surface (`--config`, `-p`/`--prompt`, `--resume`, `--dump-config`, `--dump-default-config`), whose `meta` subcommand is the same TUI over this checkout, whose `upgrade` subcommand is an option-less guided-session entry, and whose `web` subcommand is the browser UI. `src/bin.ts` switches on the resolved mode and dynamic-imports only that mode's module. `dsh --help` lists every mode and `dsh web --help` renders the web usage, `dsh --version` prints this app's version, and an unknown option or a mistyped `--resume` fails loud (stderr, exit 1) instead of misrouting. Every subcommand that shares no option with the default surface — `upgrade`, `web`, `meta` — rejects a leaked `--config`/`-p`/`--resume`/dump flag rather than running and dropping it. `dsh web`'s `--host`/`--port` are unvalidated pass-through overrides: the `dsh-host-webserver` schema is the single source of both the default (the shipped Web overlay value when a flag is absent) and validity, and rejects a bad value at boot. `--trusted-host` appends named authorities for the /api browser-trust fence; an all-interfaces bind additionally derives the machine's LAN IP literals itself ([`src/app-cli-entry.ts`](src/app-cli-entry.ts)), so the printed LAN URL works without flags.
 
 The TUI surface:
 
@@ -17,6 +17,8 @@ The TUI surface:
 
 `dsh upgrade` is a guided fresh-session entry over the default TUI surface: it mints a fresh session in the invoking directory and seeds its first turn with the bundled `dsh-upgrade` skill, exactly as if the user typed `/skill:<name>`. The launcher passes the skill name on the boot context ([`INITIAL_SKILL_KEY`](../../packages/ui/tui/README.md)), which the TUI auto-invokes once the chat is live. Both take no options — `--config`, `-p`, and `--resume` fail loud — and seed only on this first launch, so a later `dsh --resume <id>` of the session is an ordinary TUI session with no re-injection.
 
+`dsh --dump-config` and `dsh web --dump-config` print the composed config tree — the shipped base, the surface overlay, and the `--config` or personal overlay, exactly the layers that surface would boot — as YAML on stdout and exit without booting; `--dump-default-config` stops at the surface overlay, so diffing the two shows precisely what the user layer changes. Each run of rows is preceded by a `# ==` comment naming the file it comes from and the layers that patched it (e.g. `# == base.cordis.yml, patched by tui.cordis.yml`), so the output shows provenance while staying one loadable document. Composition runs through the include's own patch algorithm and YAML dialect (`applyEntryPatches`/`entryListSchema` from `@cordisjs/plugin-include`), so the dump cannot drift from what boots; `!!js` expressions print verbatim and unevaluated, and a patch whose target row is absent is reported on stderr with its layer, mirroring the Loader's boot-time warning. Launcher-owned boot-context values (session identity, CLI-flag patches) are per-invocation facts outside the config tree and do not appear. The dump flags reject boot-only flags (`-p`, `--resume`, `--config-replace`) rather than silently ignoring them, and `--dump-default-config` takes no `--config`.
+
 The Web and headless surfaces boot `base.cordis.yml` plus `web.cordis.yml`, then apply `$DSH_HOME/config.yaml`; an explicit `--config <path>` replaces that personal overlay. Both surfaces otherwise share the same composition: both treat the invoking directory as the default project and Workspace root, create named Workspaces beneath that root unless `--workspace-root <path>` overrides it, load applicable `AGENTS.md`/`CLAUDE.md` instructions into each agent-loop request prefix with a 65,536-byte render budget, opt into first-message model titles, use the same bounded transient model-request retry policy as the TUI, and mount a disposable in-memory SQLite content-index service. That service is ACTIVE at boot, while its `node:sqlite` module and database handle open only on the first content search. This keeps Node 22 startup output free of SQLite's experimental warning before search is used; the first actual search may still emit the runtime warning. Each service instance owns its database, so parallel invocations neither share unsupported SQLite state nor leave derived index files behind, and the first search lazily reconciles live and persisted logs. Headless differs only in listening on an OS-assigned port (parallel `dsh -p` runs never collide; the stderr-printed URL opens the live session in a browser). Both need the frontend dist and client bundles built (`pnpm run build && pnpm run build:web`).
 
 The shipped TUI and Web compositions register the native DeepSeek adapter plus pi-ai OpenAI and Anthropic profiles. Credentials and endpoint overrides come from the provider-standard `DEEPSEEK_API_KEY` / `DEEPSEEK_BASE_URL`, `OPENAI_API_KEY` / `OPENAI_BASE_URL`, and `ANTHROPIC_API_KEY` / `ANTHROPIC_BASE_URL` pairs in the boot's layered environment.

apps/cli/README.zh.md

@@ -3,7 +3,7 @@
 [English](README.md) | 中文
 
 
-Argv 只会通过 [Commander](https://github.com/tj/commander.js) 适配器（[`src/args.ts`](src/args.ts)）解析一次：同一个程序的默认形式（无子命令）是 TUI／无头界面（`--config`、`-p`/`--prompt`、`--resume`），`meta` 子命令是以本 checkout 为 workspace 的同一个 TUI，`upgrade` 子命令是无选项的引导会话入口，`web` 子命令则是浏览器 UI。`src/bin.ts` 按解析后的 mode 分支，仅动态导入该 mode 的模块。`dsh --help` 列出所有 mode，`dsh web --help` 渲染 Web 用法，`dsh --version` 打印此应用的版本；未知选项或拼错的 `--resume` 会明确报错（stderr，退出码 1），而不会被错路由。凡与默认界面不共享任何选项的子命令（`upgrade`、`web`、`meta`）都会拒绝泄漏进来的 `--config`/`-p`/`--resume`，而不会照常运行并丢弃它。`dsh web` 的 `--host`/`--port` 是未验证的直通覆盖：`dsh-host-webserver` schema 是默认值（标志缺失时使用已交付的 Web 覆盖层值）和有效性的唯一真源，并在启动时拒绝错误值。`--trusted-host` 为 /api 浏览器信任栅栏追加具名权威；全接口绑定还会自行推导本机的 LAN IP 字面量（[`src/app-cli-entry.ts`](src/app-cli-entry.ts)），因此打印出的 LAN URL 无需任何标志即可使用。
+Argv 只会通过 [Commander](https://github.com/tj/commander.js) 适配器（[`src/args.ts`](src/args.ts)）解析一次：同一个程序的默认形式（无子命令）是 TUI／无头界面（`--config`、`-p`/`--prompt`、`--resume`、`--dump-config`、`--dump-default-config`），`meta` 子命令是以本 checkout 为 workspace 的同一个 TUI，`upgrade` 子命令是无选项的引导会话入口，`web` 子命令则是浏览器 UI。`src/bin.ts` 按解析后的 mode 分支，仅动态导入该 mode 的模块。`dsh --help` 列出所有 mode，`dsh web --help` 渲染 Web 用法，`dsh --version` 打印此应用的版本；未知选项或拼错的 `--resume` 会明确报错（stderr，退出码 1），而不会被错路由。凡与默认界面不共享任何选项的子命令（`upgrade`、`web`、`meta`）都会拒绝泄漏进来的 `--config`/`-p`/`--resume`/dump 标志，而不会照常运行并丢弃它。`dsh web` 的 `--host`/`--port` 是未验证的直通覆盖：`dsh-host-webserver` schema 是默认值（标志缺失时使用已交付的 Web 覆盖层值）和有效性的唯一真源，并在启动时拒绝错误值。`--trusted-host` 为 /api 浏览器信任栅栏追加具名权威；全接口绑定还会自行推导本机的 LAN IP 字面量（[`src/app-cli-entry.ts`](src/app-cli-entry.ts)），因此打印出的 LAN URL 无需任何标志即可使用。
 
 TUI 界面：
 
@@ -17,6 +17,8 @@ TUI 界面：
 
 `dsh upgrade` 是默认 TUI 界面之上的引导式全新会话入口：它在调用目录中创建一个全新会话，并以内置 `dsh-upgrade` skill 播种其首轮，效果等同于用户手动键入 `/skill:<name>`。启动器将 skill 名称提供到启动上下文（[`INITIAL_SKILL_KEY`](../../packages/ui/tui/README.md)），TUI 在聊天就绪后自动调用它。两者都不接受任何选项——`--config`、`-p`、`--resume` 都会明确报错——且仅在首次启动时播种，因此之后 `dsh --resume <id>` 恢复该会话时是普通 TUI 会话，不会重复注入。
 
+`dsh --dump-config` 和 `dsh web --dump-config` 把合成后的配置树——已交付的基础配置、界面覆盖层，以及 `--config` 或个人覆盖层，恰好是该界面启动时组装的那些层——以 YAML 打印到 stdout 后退出，不启动任何东西；`--dump-default-config` 止步于界面覆盖层，因此对两份输出做 diff 就能精确看出用户层改了什么。每段连续的行之前都有一条 `# ==` 注释，标明该段来自哪个文件以及被哪些层修补过（例如 `# == base.cordis.yml, patched by tui.cordis.yml`），因此输出既展示来源，又仍是一份可加载的文档。合成通过 include 自己的补丁算法和 YAML 方言（`@cordisjs/plugin-include` 的 `applyEntryPatches`/`entryListSchema`）完成，因此 dump 不可能与实际启动漂移；`!!js` 表达式原样打印、不求值，目标行不存在的补丁会连同其所在层报到 stderr，与 Loader 启动时的警告一致。由启动器持有的启动上下文值（会话身份、CLI 标志补丁）是每次调用的事实，位于配置树之外，不会出现。dump 标志会拒绝仅用于启动的标志（`-p`、`--resume`、`--config-replace`）而不是静默忽略它们，`--dump-default-config` 不接受 `--config`。
+
 Web 和无头界面启动 `base.cordis.yml` 与 `web.cordis.yml`，随后应用 `$DSH_HOME/config.yaml`；显式的 `--config <path>` 会替代该个人覆盖。除此之外，两者共享同一套组合：两者都将调用目录视为默认项目和 Workspace 根目录，除非通过 `--workspace-root <path>` 覆盖，否则会在该根目录下创建具名 Workspace；它们会把适用的 `AGENTS.md`/`CLAUDE.md` 指令加载到每个 agent-loop 请求前缀中，渲染预算为 65,536 字节，选用首条消息模型标题，采用与 TUI 相同的有界暂时性模型请求重试策略，并挂载一个可丢弃的内存 SQLite 内容索引服务。该服务在启动时处于 ACTIVE 状态，但其 `node:sqlite` 模块与数据库句柄分别要到首次内容搜索才会导入和打开。这样可使 Node 22 在尚未使用搜索时的启动输出不出现 SQLite 实验性警告；首次实际搜索仍可能发出运行时警告。每个服务实例独占自己的数据库，因此并行调用既不会共享不受支持的 SQLite 状态，也不会留下派生索引文件，首次搜索还会惰性对账实时日志与持久化日志。无头界面唯一的差异是监听操作系统分配的端口（并行 `dsh -p` 运行绝不冲突；stderr 打印的 URL 会在浏览器中打开实时会话）。两者都需要先构建前端 dist 和客户端 bundle（`pnpm run build && pnpm run build:web`）。
 
 已交付的 TUI 和 Web 组合会注册原生 DeepSeek 适配器，以及 pi-ai 的 OpenAI 和 Anthropic 提供方配置。凭据和端点覆盖来自启动分层环境中的提供方标准变量对：`DEEPSEEK_API_KEY` / `DEEPSEEK_BASE_URL`、`OPENAI_API_KEY` / `OPENAI_BASE_URL` 和 `ANTHROPIC_API_KEY` / `ANTHROPIC_BASE_URL`。

apps/cli/src/args.ts

@@ -23,6 +23,22 @@ interface TuiInvocation {
   resume?: string
 }
 
+/**
+ * Print the composed config tree and exit, without booting: `--dump-config`
+ * composes the shipped base, the surface overlay, and the `--config` or
+ * personal overlay — exactly the layers that surface would boot;
+ * `--dump-default-config` stops at the surface overlay (the shipped tree, no
+ * user layer).
+ */
+interface DumpConfigInvocation {
+  mode: 'dump-config'
+  surface: 'tui' | 'web'
+  /** Omit the `--config`/personal layer and print only the shipped composition. */
+  defaultOnly: boolean
+  /** The `--config` overlay to compose instead of the personal one. */
+  config?: string
+}
+
 /** Headless one-shot: `dsh -p "task"`. */
 interface HeadlessInvocation {
   mode: 'headless'
@@ -69,6 +85,7 @@ interface WebInvocation {
 /** The resolved `dsh` invocation: exactly one mode. `--help`/`--version`/errors exit inside {@link parseDshArgs}. */
 export type DshInvocation =
   | TuiInvocation
+  | DumpConfigInvocation
   | HeadlessInvocation
   | MetaInvocation
   | SkillSessionInvocation
@@ -82,6 +99,34 @@ interface WebOptions {
   dev?: boolean
   workspaceRoot?: string
   trustedHost?: string[]
+  dumpConfig?: boolean
+  dumpDefaultConfig?: boolean
+}
+
+/**
+ * Resolve the two dump flags for one surface, or return `undefined` when
+ * neither was passed. Both flags together are contradictory (one includes the
+ * user layer, the other excludes it) and fail loud through `error`.
+ */
+function resolveDump(
+  surface: 'tui' | 'web',
+  options: { config?: string; dumpConfig?: boolean; dumpDefaultConfig?: boolean },
+  error: (message: string) => never,
+): DumpConfigInvocation | undefined {
+  if (options.dumpConfig !== true && options.dumpDefaultConfig !== true) return undefined
+  if (options.dumpConfig === true && options.dumpDefaultConfig === true) {
+    error('error: --dump-config and --dump-default-config are mutually exclusive')
+  }
+  const defaultOnly = options.dumpDefaultConfig === true
+  if (defaultOnly && options.config !== undefined) {
+    error('error: --dump-default-config prints the shipped tree and takes no --config')
+  }
+  return {
+    mode: 'dump-config',
+    surface,
+    defaultOnly,
+    ...options.config !== undefined && { config: options.config },
+  }
 }
 
 /**
@@ -135,7 +180,26 @@ Examples:
     .option('--resume <id>', 'continue a past session by id')
     .option('--config <path>', 'apply this overlay of loader patches instead of the personal one')
     .option('--config-replace <path>', 'boot this file as the entire tree, ignoring the shipped and personal configuration')
-    .action((options: { config?: string; configReplace?: string; prompt?: string; resume?: string }) => {
+    .option('--dump-config', 'print the composed config tree (base + surface + --config/personal overlay) and exit')
+    .option('--dump-default-config', 'print the shipped config tree (base + surface overlay, no user layer) and exit')
+    .action((options: {
+      config?: string
+      configReplace?: string
+      prompt?: string
+      resume?: string
+      dumpConfig?: boolean
+      dumpDefaultConfig?: boolean
+    }) => {
+      const dump = resolveDump('tui', options, message => program.error(message))
+      if (dump !== undefined) {
+        // The dump prints composition; a boot-only flag alongside it would be
+        // silently ignored, so reject the mix loud.
+        if (options.prompt !== undefined || options.resume !== undefined || options.configReplace !== undefined) {
+          program.error('error: --dump-config/--dump-default-config take none of -p/--prompt, --resume, or --config-replace')
+        }
+        resolved = dump
+        return
+      }
       if (options.prompt !== undefined) {
         // A headless prompt owns the invocation; an empty task has nothing to
         // run, and --config/--resume are TUI inputs that must not silently
@@ -168,10 +232,18 @@ Examples:
   // a leaked config/prompt/resume option is a mistyped invocation that must fail
   // loud rather than silently run and drop the input.
   const rejectParentOptions = (command: string): void => {
-    const parent = program.opts<{ config?: string; configReplace?: string; prompt?: string; resume?: string }>()
+    const parent = program.opts<{
+      config?: string
+      configReplace?: string
+      prompt?: string
+      resume?: string
+      dumpConfig?: boolean
+      dumpDefaultConfig?: boolean
+    }>()
     if (parent.config !== undefined || parent.configReplace !== undefined
-      || parent.prompt !== undefined || parent.resume !== undefined) {
-      program.error(`error: ${command} takes none of --config, --config-replace, -p/--prompt, or --resume`)
+      || parent.prompt !== undefined || parent.resume !== undefined
+      || parent.dumpConfig !== undefined || parent.dumpDefaultConfig !== undefined) {
+      program.error(`error: ${command} takes none of --config, --config-replace, -p/--prompt, --resume, --dump-config, or --dump-default-config`)
     }
   }
 
@@ -198,8 +270,15 @@ Examples:
     .option('--dev', 'developer mode: hot-reload the browser client')
     .option('--workspace-root <path>', 'parent directory for workspaces created from the browser UI')
     .option('--trusted-host <authority...>', 'extra authority the /api browser-trust fence accepts (host or host:port; repeatable)')
+    .option('--dump-config', 'print the composed config tree (base + web + --config/personal overlay) and exit')
+    .option('--dump-default-config', 'print the shipped config tree (base + web overlay, no user layer) and exit')
     .action((options: WebOptions) => {
       rejectParentOptions('web')
+      const dump = resolveDump('web', options, message => program.error(message))
+      if (dump !== undefined) {
+        resolved = dump
+        return
+      }
       resolved = resolveWeb(options)
     })
 

apps/cli/src/bin.ts

@@ -43,6 +43,11 @@ switch (invocation.mode) {
     await runTui(invocation.config, invocation.resume, undefined, undefined, invocation.configReplace)
     break
   }
+  case 'dump-config': {
+    const { runDumpConfig } = await import('./dump-config.ts')
+    runDumpConfig(invocation.surface, invocation.defaultOnly, invocation.config)
+    break
+  }
   case 'meta': {
     const { runMeta } = await import('./tui.ts')
     await runMeta()

apps/cli/src/dump-config.ts

@@ -0,0 +1,61 @@
+/**
+ * `dsh --dump-config` / `dsh web --dump-config` — print the composed config
+ * tree without booting: the shipped base, the surface overlay, and (unless
+ * `--dump-default-config`) the `--config` or personal overlay, composed
+ * through the include's own patch algorithm so the printed tree is exactly
+ * what that surface would mount. `!!js` expressions print verbatim,
+ * unevaluated — the dump shows composition, not one process's environment.
+ * Launcher-provided boot-context values (session identity, CLI-flag patches)
+ * are per-invocation facts outside the config tree and do not appear.
+ * @module @deepseek-ai/dsh/dump-config
+ */
+
+import { basename, join } from 'node:path'
+import { fileURLToPath } from 'node:url'
+import {
+  loadOverlayPatches,
+  loadPersonalPatches,
+  PERSONAL_CONFIG_FILENAME,
+  renderConfigDump,
+  type ConfigDumpLayer,
+} from '@deepseek-ai/dsh-app-boot'
+import { resolveDshHome } from '@deepseek-ai/dsh-paths'
+
+const NAME = 'dsh'
+
+const BASE_CONFIG = fileURLToPath(new URL('../config/base.cordis.yml', import.meta.url))
+const SURFACE_OVERLAYS = {
+  tui: fileURLToPath(new URL('../config/tui.cordis.yml', import.meta.url)),
+  web: fileURLToPath(new URL('../config/web.cordis.yml', import.meta.url)),
+} as const
+
+/* v8 ignore start -- composition over the unit-tested renderConfigDump; the
+   built-bin e2e drives this path end to end */
+/**
+ * Print one surface's composed config tree to stdout, with a comment
+ * separator naming the file each section of rows comes from (and the layers
+ * that patched it).
+ * @param surface - which surface overlay to compose over the shared base.
+ * @param defaultOnly - stop at the surface overlay (no `--config`/personal layer).
+ * @param config - the `--config` overlay path composed instead of the personal
+ * one, or `undefined` to use `$DSH_HOME/config.yaml`.
+ */
+export function runDumpConfig(surface: 'tui' | 'web', defaultOnly: boolean, config?: string): void {
+  const overlay = SURFACE_OVERLAYS[surface]
+  const layers: ConfigDumpLayer[] = [
+    { label: basename(overlay), patches: loadOverlayPatches(NAME, overlay) },
+  ]
+  if (!defaultOnly) {
+    if (config === undefined) {
+      const personal = loadPersonalPatches(NAME)
+      // The personal file may be absent; the shipped layers still print.
+      if (personal !== undefined) {
+        layers.push({ label: join(resolveDshHome(), PERSONAL_CONFIG_FILENAME), patches: personal })
+      }
+    } else {
+      layers.push({ label: config, patches: loadOverlayPatches(NAME, config) })
+    }
+  }
+  process.stdout.write(renderConfigDump(NAME, BASE_CONFIG, layers))
+}
+/* v8 ignore stop */

apps/cli/tests/args.spec.ts

@@ -45,6 +45,29 @@ describe('parseDshArgs', () => {
       .toEqual({ mode: 'web', dev: false, trustedHosts: ['harness.internal:3080', 'lab.internal', '10.0.0.9'] })
   })
 
+  it('routes the dump flags per surface: composed with the user layer, or shipped only', () => {
+    expect(parse(['--dump-config'])).toEqual({ mode: 'dump-config', surface: 'tui', defaultOnly: false })
+    expect(parse(['--dump-config', '--config', 'c.yml']))
+      .toEqual({ mode: 'dump-config', surface: 'tui', defaultOnly: false, config: 'c.yml' })
+    expect(parse(['--dump-default-config'])).toEqual({ mode: 'dump-config', surface: 'tui', defaultOnly: true })
+    expect(parse(['web', '--dump-config'])).toEqual({ mode: 'dump-config', surface: 'web', defaultOnly: false })
+    expect(parse(['web', '--dump-config', '--config', 'w.yml']))
+      .toEqual({ mode: 'dump-config', surface: 'web', defaultOnly: false, config: 'w.yml' })
+    expect(parse(['web', '--dump-default-config'])).toEqual({ mode: 'dump-config', surface: 'web', defaultOnly: true })
+    // The two dump flags contradict each other; boot-only flags alongside a
+    // dump would be silently ignored; the shipped tree takes no user overlay.
+    expect(exitCode(['--dump-config', '--dump-default-config'])).toBe(1)
+    expect(exitCode(['--dump-default-config', '--config', 'c.yml'])).toBe(1)
+    expect(exitCode(['--dump-config', '--resume', 's'])).toBe(1)
+    expect(exitCode(['--dump-config', '-p', 'task'])).toBe(1)
+    expect(exitCode(['--dump-config', '--config-replace', 'tree.yml'])).toBe(1)
+    expect(exitCode(['web', '--dump-config', '--dump-default-config'])).toBe(1)
+    expect(exitCode(['web', '--dump-default-config', '--config', 'w.yml'])).toBe(1)
+    // A leaked dump flag on a subcommand that has none is a mistyped invocation.
+    expect(exitCode(['meta', '--dump-config'])).toBe(1)
+    expect(exitCode(['upgrade', '--dump-config'])).toBe(1)
+  })
+
   it('exits nonzero instead of silently starting fresh or dropping inputs', () => {
     // Empty resume/prompt would be swallowed downstream; --prompt mixed with
     // TUI inputs must not lose them. (Bad host/port are gated by the webserver

apps/cli/tests/built-bin.e2e.ts

@@ -1,8 +1,9 @@
-import { existsSync } from 'node:fs'
+import { existsSync, mkdtempSync, rmSync, writeFileSync } from 'node:fs'
+import { tmpdir } from 'node:os'
 import { join } from 'node:path'
 import { fileURLToPath } from 'node:url'
 import { execa } from 'execa'
-import { describe, expect, it } from 'vitest'
+import { afterEach, beforeEach, describe, expect, it } from 'vitest'
 
 /**
  * Published-entry smoke for the `dsh` bin: run the built `lib/bin.js` under
@@ -22,13 +23,20 @@ import { describe, expect, it } from 'vitest'
 const repoRoot = fileURLToPath(new URL('../../../', import.meta.url))
 const dshBin = join(repoRoot, 'apps/cli/lib/bin.js')
 
-/** Run the built bin with PIPED stdio (stdin closed at EOF); resolve with output + exit code. */
-async function runBuiltBin(): Promise<{ stdout: string; code: number; stderr: string }> {
-  const result = await execa(process.execPath, [dshBin], {
+/**
+ * Run the built bin with PIPED stdio (stdin closed at EOF); resolve with output
+ * + exit code. `env` isolates the Harness home for surfaces that read it.
+ */
+async function runBuiltBin(
+  args: readonly string[] = [],
+  env: Record<string, string> = {},
+): Promise<{ stdout: string; code: number; stderr: string }> {
+  const result = await execa(process.execPath, [dshBin, ...args], {
     input: '',
     timeout: 25_000,
     killSignal: 'SIGKILL',
     reject: false,
+    env,
   })
   if (result.timedOut) {
     throw new Error(`dsh built bin did not exit within 25s. stdout:\n${result.stdout}\nstderr:\n${result.stderr}`)
@@ -45,4 +53,61 @@ describe.skipIf(!existsSync(dshBin))('dsh BUILT bin (node lib/bin.js, no tsx)',
     // The refusal happens before any plugin mounts: stdout stays silent.
     expect(stdout).toBe('')
   }, 30_000)
+
+  describe('dsh --dump-config', () => {
+    let home: string
+    beforeEach(() => { home = mkdtempSync(join(tmpdir(), 'dsh-dump-bin-')) })
+    afterEach(() => { rmSync(home, { recursive: true, force: true }) })
+
+    it('prints the shipped TUI composition without booting or needing a TTY', async () => {
+      const { stdout, code, stderr } = await runBuiltBin(['--dump-default-config'], { DSH_HOME: home })
+      expect(code).toBe(0)
+      expect(stderr).toBe('')
+      // Base rows composed with the TUI overlay's surface values, `!!js`
+      // expressions verbatim (unevaluated), and TUI-only inserted rows present.
+      expect(stdout).toContain("name: '@deepseek-ai/dsh-agent-loop'")
+      expect(stdout).toContain('model: deepseek-v4-pro')
+      expect(stdout).toContain('cwd: !!js process.cwd()')
+      expect(stdout).toContain("name: '@deepseek-ai/dsh-tui'")
+      // Provenance comment separators name each section's source file.
+      expect(stdout).toContain('# == base.cordis.yml')
+      expect(stdout).toContain('# == base.cordis.yml, patched by tui.cordis.yml')
+      expect(stdout).toContain('# == tui.cordis.yml')
+    }, 30_000)
+
+    it('layers the personal overlay in --dump-config and reports an unmatched patch on stderr', async () => {
+      writeFileSync(join(home, 'config.yaml'), [
+        '- id: agent-loop',
+        '  config:',
+        '    agents:',
+        '      - id: main',
+        '        provider: custom-provider',
+        '        model: custom-model',
+        '- id: only-on-web',
+        '  config:',
+        '    value: 1',
+        '',
+      ].join('\n'))
+      const { stdout, code, stderr } = await runBuiltBin(['--dump-config'], { DSH_HOME: home })
+      expect(code).toBe(0)
+      expect(stdout).toContain('provider: custom-provider')
+      expect(stdout).not.toContain('model: deepseek-v4-pro')
+      // The personal layer appears in the patched row's provenance and the
+      // skipped-patch warning carries its label.
+      expect(stdout).toContain(`patched by tui.cordis.yml, ${join(home, 'config.yaml')}`)
+      expect(stderr).toContain('patch: entry "only-on-web" not found')
+
+      // The shipped view ignores the personal overlay entirely.
+      const shipped = await runBuiltBin(['--dump-default-config'], { DSH_HOME: home })
+      expect(shipped.stdout).not.toContain('custom-provider')
+      expect(shipped.stdout).toContain('model: deepseek-v4-pro')
+    }, 30_000)
+
+    it('composes the web overlay for `dsh web --dump-config`', async () => {
+      const { stdout, code } = await runBuiltBin(['web', '--dump-config'], { DSH_HOME: home })
+      expect(code).toBe(0)
+      expect(stdout).toContain("name: '@deepseek-ai/dsh-host-webserver'")
+      expect(stdout).not.toContain("name: '@deepseek-ai/dsh-tui'")
+    }, 30_000)
+  })
 })

apps/web/tests/access-confirmation.e2e.ts

@@ -11,7 +11,7 @@ import {
   assertFixtureInventory, captureStableAria, compareOrRefreshGolden,
   launchWebScaffold, watchConsole, webSnapshotMode, type WebScaffold,
 } from './scaffold.ts'
-import { saveFailureShot } from './support.ts'
+import { ZH_BROWSER_LOCALE, saveFailureShot } from './support.ts'
 
 /**
  * connectFreshWorkspace twin over the product default Chinese locale (the
@@ -49,7 +49,7 @@ describe('web e2e: Full access confirmation', () => {
     browser = await chromium.launch(executablePath === undefined ? {} : { executablePath })
     // Keep the product default Chinese locale: the golden pins the actual
     // registered dictionary rather than a test-local translation callback.
-    page = await browser.newPage({ viewport: { width: 1680, height: 1000 } })
+    page = await browser.newPage({ viewport: { width: 1680, height: 1000 }, locale: ZH_BROWSER_LOCALE })
     tripwire = watchConsole(page)
     await page.goto(scaffold.baseUrl, { waitUntil: 'load' })
     await page.waitForSelector('[class*="frame"]', { timeout: 30_000 })

apps/web/tests/message-actions.e2e.ts

@@ -64,14 +64,14 @@ describe('web e2e: message IconActions and clocks on settled history', () => {
     await expect.poll(() => page.getByText('DONE', { exact: true }).count(), { timeout: 15_000 }).toBe(1)
 
     // Focus-reveal the footers (hover:hover keeps them opacity-hidden until
-    // hover/focus-within). User has three actions; each turn's last content
-    // assistant has copy + branch.
+    // hover/focus-within). User and each turn's last content assistant both
+    // have copy + branch.
     const copyButtons = page.getByRole('button', { name: 'Copy' })
     await expect.poll(() => copyButtons.count(), { timeout: 10_000 }).toBeGreaterThanOrEqual(2)
     await copyButtons.first().focus()
     await expect.poll(() => page.getByRole('button', { name: 'Branch into a new conversation' }).count(), { timeout: 5_000 })
       .toBeGreaterThanOrEqual(2)
-    await expect.poll(() => page.getByRole('button', { name: 'Edit' }).count(), { timeout: 5_000 }).toBe(1)
+    await expect.poll(() => page.getByRole('button', { name: 'Edit' }).count(), { timeout: 5_000 }).toBe(0)
   }, 60_000)
 
   it.skipIf(MODE === 'record')('matches the conversation aria golden with IconActions and clocks', async () => {

apps/web/tests/models-settings.e2e.ts

@@ -20,7 +20,7 @@ import {
   assertFixtureInventory, captureStableAria, compareOrRefreshGolden,
   launchWebScaffold, watchConsole, webSnapshotMode, type WebScaffold,
 } from './scaffold.ts'
-import { saveFailureShot } from './support.ts'
+import { ZH_BROWSER_LOCALE, saveFailureShot } from './support.ts'
 
 const SNAPSHOT_DIR = fileURLToPath(new URL('./snapshots/models-settings', import.meta.url))
 const EMPTY_EXPECTED = join(SNAPSHOT_DIR, 'empty.expected.md')
@@ -37,7 +37,8 @@ describe('web e2e: Models settings page configures a dormant provider', () => {
   beforeAll(async () => {
     scaffold = await launchWebScaffold({})
     browser = await chromium.launch()
-    page = await browser.newPage({ viewport: { width: 1680, height: 1000 } })
+    // The scenario asserts the shipped Chinese copy, so the browser asks for it.
+    page = await browser.newPage({ viewport: { width: 1680, height: 1000 }, locale: ZH_BROWSER_LOCALE })
     tripwire = watchConsole(page)
     await page.goto(scaffold.baseUrl, { waitUntil: 'load' })
     await page.waitForSelector('[class*="frame"]', { timeout: 30_000 })

apps/web/tests/onboarding-deepseek-config.e2e.ts

@@ -12,7 +12,7 @@ import {
   acknowledgeReloadConnectionLoss, assertFixtureInventory, captureStableAria, compareOrRefreshGolden,
   launchWebScaffold, watchConsole, webSnapshotMode, type WebScaffold,
 } from './scaffold.ts'
-import { saveFailureShot } from './support.ts'
+import { ZH_BROWSER_LOCALE, saveFailureShot } from './support.ts'
 import { settingsNamespace } from '@deepseek-ai/dsh-settings'
 import {
   WELCOME_NOTICE_ACK_FIELD, WELCOME_NOTICE_COPY, WELCOME_NOTICE_SETTINGS_NAMESPACE,
@@ -34,7 +34,8 @@ describe.skipIf(MODE === 'record')('web e2e: first-run DeepSeek credential setup
   beforeAll(async () => {
     scaffold = await launchWebScaffold({ deepSeekMissingCredential: true, welcomeNoticePending: true })
     browser = await chromium.launch()
-    page = await browser.newPage({ viewport: { width: 1440, height: 960 } })
+    // The scenario asserts the shipped Chinese copy, so the browser asks for it.
+    page = await browser.newPage({ viewport: { width: 1440, height: 960 }, locale: ZH_BROWSER_LOCALE })
     tripwire = watchConsole(page)
     page.on('console', message => browserConsole.push(message.text()))
     await page.goto(scaffold.baseUrl, { waitUntil: 'load' })

apps/web/tests/permission-policy-context.e2e.ts

@@ -0,0 +1,171 @@
+// Web acceptance for current sandbox-policy context. A real Chromium drives
+// the shipped /permission command through all three presets; record mode uses
+// the real provider, while replay keeps the same provider-authored behavior
+// keyless. Assertions read the exact durable header, runtime-context messages,
+// and tool calls, so assistant prose alone cannot satisfy the scenario.
+import { readFile } from 'node:fs/promises'
+import { join } from 'node:path'
+import { fileURLToPath } from 'node:url'
+import type { Browser, Page } from 'playwright'
+import { chromium } from 'playwright'
+import { afterAll, beforeAll, describe, expect, it, onTestFailed } from 'vitest'
+import { canonicalPath } from '@deepseek-ai/dsh-sandbox'
+import type { SessionEvent } from '@deepseek-ai/dsh-session'
+import {
+  assertFixtureInventory, fixtureUserPrompts, launchWebScaffold, recordFixture,
+  watchConsole, webSnapshotMode, type WebScaffold,
+} from './scaffold.ts'
+import { connectFreshWorkspace, newEnglishPage, saveFailureShot } from './support.ts'
+
+const SNAPSHOT_DIR = fileURLToPath(new URL('./snapshots/permission-policy-context', import.meta.url))
+const FIXTURE = fileURLToPath(new URL('./snapshots/permission-policy-context/session.jsonl', import.meta.url))
+const MODE = webSnapshotMode()
+
+const PROMPTS = [
+  'Can you create or edit a normal file right now under the current policy? Answer directly in one sentence. Do not call a tool just to discover the policy.',
+  'Does the DSH file sandbox currently restrict file operations? Answer directly in one sentence. Do not call tools.',
+  'Reply with exactly WORKSPACE_POLICY_SEEN. Do not call tools.',
+  'Create the relative path policy-neutral.txt in the current workspace containing exactly POLICY_NEUTRAL_OK, verify its contents, then report completion.',
+] as const
+
+const PRESET_LABELS = ['Read Only', 'Full access', 'Workspace Write'] as const
+
+function requestSystems(events: readonly SessionEvent[]): string[] {
+  return events.flatMap((event) => {
+    if (event.type !== 'request/header') return []
+    return typeof event.data.header.system === 'string' ? [event.data.header.system] : []
+  })
+}
+
+function runtimeContexts(events: readonly SessionEvent[]): string[] {
+  return events.flatMap((event) => {
+    if (event.type !== 'user/message'
+      || event.data.source.kind !== 'plugin'
+      || event.data.source.plugin !== '@deepseek-ai/dsh-system-prompt') return []
+    return event.data.content.flatMap(block => block.type === 'text' ? [block.text] : [])
+  })
+}
+
+function assistantTexts(events: readonly SessionEvent[]): string[] {
+  return events.flatMap((event) => {
+    if (event.type !== 'assistant/message') return []
+    const text = event.data.message.content.flatMap(block => block.type === 'text' ? [block.text] : []).join('').replaceAll('**', '')
+    return text.length === 0 ? [] : [text]
+  })
+}
+
+function callArgs(event: Extract<SessionEvent, { type: 'tool/call' }>): Record<string, unknown> {
+  return JSON.parse(event.data.arguments) as Record<string, unknown>
+}
+
+describe('web e2e: current sandbox policy reaches the model before tools', () => {
+  let scaffold: WebScaffold
+  let browser: Browser
+  let page: Page
+  let tripwire: ReturnType<typeof watchConsole>
+  let disposeApproval: (() => void) | undefined
+  let sessionWorkspace: string | undefined
+  const sessionEvents: SessionEvent[] = []
+
+  beforeAll(async () => {
+    scaffold = await launchWebScaffold(MODE === 'record' ? {} : { replayFixture: FIXTURE })
+    disposeApproval = scaffold.ctx.on('approval/request', () => Promise.resolve('allowed-once'), { prepend: true })
+    scaffold.ctx.on('session/event', (session, event: SessionEvent) => {
+      sessionWorkspace = session.header.cwd
+      sessionEvents.push(event)
+    })
+    browser = await chromium.launch()
+    page = await newEnglishPage(browser)
+    tripwire = watchConsole(page)
+    await page.goto(scaffold.baseUrl, { waitUntil: 'load' })
+    await page.waitForSelector('[class*="frame"]', { timeout: 30_000 })
+    await connectFreshWorkspace(page)
+  }, 120_000)
+
+  afterAll(async () => {
+    await browser?.close()
+    disposeApproval?.()
+    await scaffold?.close()
+  })
+
+  it('switches read-only, danger-full-access, and workspace-write through the real GUI command path', async () => {
+    onTestFailed(() => saveFailureShot(page, 'web-e2e-permission-policy-context'))
+    if (MODE !== 'record') {
+      expect(fixtureUserPrompts(await readFile(FIXTURE, 'utf8'))).toEqual(PROMPTS)
+    }
+
+    const input = page.locator('textarea').first()
+    let sessionId: Awaited<ReturnType<WebScaffold['whenTurnSettled']>> | undefined
+    for (const [index, preset] of ['read-only', 'danger-full-access', 'workspace-write'].entries()) {
+      await input.fill(`/permission ${preset}`)
+      await input.press('Enter')
+      await page.getByRole('button', { name: `Access mode, current: ${PRESET_LABELS[index]}` })
+        .waitFor({ timeout: 10_000 })
+
+      const settled = scaffold.whenTurnSettled()
+      await input.fill(PROMPTS[index] as string)
+      await input.press('Enter')
+      sessionId = await settled
+      await expect.poll(() => input.isEnabled(), { timeout: 10_000 }).toBe(true)
+    }
+
+    await input.fill('/permission read-only')
+    await input.press('Enter')
+    await page.getByRole('button', { name: 'Access mode, current: Read Only' }).waitFor({ timeout: 10_000 })
+    const settled = scaffold.whenTurnSettled()
+    await input.fill(PROMPTS[3])
+    await input.press('Enter')
+    sessionId = await settled
+
+    if (sessionId === undefined) throw new Error('permission-policy scenario completed no model turn')
+    if (MODE === 'record') await recordFixture(scaffold, sessionId, FIXTURE)
+  }, 240_000)
+
+  it.skipIf(MODE === 'record')('records cache-safe current policy before the corresponding model behavior', async () => {
+    const systems = requestSystems(sessionEvents)
+    expect(systems).toHaveLength(1)
+    expect(systems[0]).not.toContain('Current DSH file policy:')
+    expect(systems[0]).not.toContain('Approval policy:')
+    expect(systems[0]).not.toContain('Approval prompts are disabled in this session')
+
+    const contexts = runtimeContexts(sessionEvents)
+    expect(contexts).toHaveLength(4)
+    expect(contexts[0]).toContain('Current DSH file policy: read-only. Any available operation enforced by the DSH file sandbox cannot modify files in the standing mode.')
+    expect(contexts[0]).toContain('Do not refuse a required modification from this policy alone')
+    expect(contexts[0]).toContain('Approval policy: ask.')
+    expect(contexts[1]).toContain('Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.')
+    expect(contexts[1]).toContain('Approval prompts are disabled in this session')
+
+    if (sessionWorkspace === undefined) throw new Error('permission-policy scenario observed no session workspace')
+    expect(contexts[2]).toContain(`Current DSH file policy: workspace-write. Any available operation enforced by the DSH file sandbox may modify files under the session workspace: ${JSON.stringify(canonicalPath(sessionWorkspace))}. Some platform temporary areas may also be writable.`)
+    expect(contexts[2]).toContain('Approval policy: ask.')
+    expect(contexts[2]).not.toContain('Approval prompts are disabled in this session')
+    expect(contexts[3]).toContain('Current DSH file policy: read-only.')
+
+    const answers = assistantTexts(sessionEvents)
+    expect(answers.length).toBeGreaterThanOrEqual(4)
+    expect(answers[0]).toMatch(/read-only.*(?:denied|cannot modify|cannot create or edit)/i)
+    expect(answers[1]).toMatch(/does not restrict.*(?:file operations|(?:write\/edit tools|write and edit tools).*one-shot bash commands)/i)
+    expect(answers[2]).toBe('WORKSPACE_POLICY_SEEN')
+    const calls = sessionEvents.filter(
+      (event): event is Extract<SessionEvent, { type: 'tool/call' }> => event.type === 'tool/call',
+    )
+    expect(calls.every(call => call.data.turn === 4)).toBe(true)
+    expect(calls.length).toBeGreaterThanOrEqual(2)
+    const firstCall = calls[0]
+    if (firstCall === undefined) throw new Error('neutral policy task produced no tool call')
+    expect(callArgs(firstCall)['sandbox_permissions']).toBeUndefined()
+    expect(calls.some(call => callArgs(call)['sandbox_permissions'] !== undefined)).toBe(true)
+    expect(sessionEvents.some(event => event.type === 'tool/result'
+      && JSON.stringify(event.data).includes('[sandbox: file access denied under read-only mode]'))).toBe(true)
+    expect(sessionEvents.some(event => event.type === 'approval/asked')).toBe(true)
+    if (sessionWorkspace === undefined) throw new Error('permission-policy scenario observed no session workspace')
+    expect(await readFile(join(sessionWorkspace, 'policy-neutral.txt'), 'utf8')).toBe('POLICY_NEUTRAL_OK')
+  })
+
+  it.skipIf(MODE === 'record')('stays clean and keeps the fixture inventory closed', async () => {
+    expect(tripwire.pageErrors).toEqual([])
+    expect(tripwire.warnings).toEqual([])
+    await assertFixtureInventory(SNAPSHOT_DIR, ['session.jsonl'])
+  })
+})

apps/web/tests/queue-actions.e2e.ts

@@ -131,7 +131,7 @@ describe('web e2e: queue row actions', () => {
 
     const snapshot = await captureStableAria(page, '[class*="centerCol"]', scaffold.workspaceCwd)
     await compareOrRefreshGolden(UI_EXPECTED, snapshot, MODE)
-    expect(sessionEvents.filter(event => event.type === 'user/message')).toHaveLength(1)
+    expect(sessionEvents.filter(event => event.type === 'user/message' && event.data.source.kind === 'user')).toHaveLength(1)
     expect(tripwire.pageErrors).toEqual([])
     expect(tripwire.warnings).toEqual([])
 

apps/web/tests/scaffold.ts

@@ -11,13 +11,14 @@
 // masking its credential, without making a model call.
 //
 // Composition divergences from `dsh web`, all deliberate, all via include
-// patches after the shipped surface overlay: temp persistenceRoot; local skill
-// roots confined to the temp workspace; workspace-context disabled (recorded
-// fixtures must not embed this repo's AGENTS.md); session-title-llm disabled
-// (its fire-and-forget title call would race the loop for the session's replay
-// cursor); webserver pinned to port 0 with the built dist; ordinary keyless
-// modes disable llm-deepseek and fill the open llm seam post-boot with
-// installLlmReplay on the settled root ctx
+// patches after the shipped surface overlay, over the SAME tree (never a
+// second yml): temp persistenceRoot; host-level skill roots confined to the
+// temp workspace while project skill discovery remains real; workspace-context
+// disabled (recorded fixtures must not embed this repo's AGENTS.md);
+// session-title-llm disabled (its fire-and-forget title call would race the
+// loop for the session's replay cursor); webserver pinned to port 0 with the
+// built dist; ordinary keyless modes disable llm-deepseek and fill the open
+// llm seam post-boot with installLlmReplay on the settled root ctx
 // (the plugin-row path discards the ReplayHandle; the direct install keeps
 // assertConsumed for the teardown fixture-consumption check).
 import { existsSync } from 'node:fs'

apps/web/tests/search-card.snapshot.ts

@@ -0,0 +1,164 @@
+// @vitest-environment jsdom
+// Assembled search-card snapshot: boots the real built `packages/client/*/lib/
+// client.js` bundles through AppWebEntry's ModuleLoader path against the keyless
+// FixtureApiClient transport (no API key, no model round), opens the fixture
+// session, and pins the search card the `grep` turn (fixture turn 66) renders in
+// the assembled application. The built-boot smoke proves the graph boots but
+// carries no behavior assertions by contract; this is the assembled-output check
+// that a broken SearchRow registration or a dropped card would fail — the
+// per-package suites bench over src and cannot see the bundled wiring.
+//
+// Keyless and deterministic: the fixture is the fake server, so the grep turn's
+// matches, its truncation summary, and its head/tail cap are fixed in the
+// fixture, not harvested from a live model. The recovery-footer arm is a pure
+// derivation over the result view, pinned at every render site by the
+// ui-conversation suite; here the fixture turn exercises the assembled card
+// shape and its cap.
+import { mkdirSync, readFileSync, writeFileSync } from 'node:fs'
+import { dirname, join } from 'node:path'
+import { act, cleanup, fireEvent, screen, waitFor, within } from '@testing-library/react'
+import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest'
+import type { WebBootEntry } from '@deepseek-ai/dsh-client-modules/client'
+import { AppWebEntry } from '@deepseek-ai/dsh-client-web'
+
+const EXPECTED = join(process.cwd(), 'apps/web/tests/snapshots/search-card/grep-card.expected.txt')
+const refreshing = process.env.DSH_SNAPSHOT === 'record' || process.env.DSH_SNAPSHOT === 'refresh'
+
+const PLUGINS: readonly (WebBootEntry & { dir: string })[] = [
+  { id: '@deepseek-ai/dsh-client-connection', dir: 'connection', url: '/plugins/connection.js', rev: 'fx', inject: [], immediately: true },
+  { id: '@deepseek-ai/dsh-client-runtime', dir: 'runtime', url: '/plugins/runtime.js', rev: 'fx', inject: ['@deepseek-ai/dsh-client-connection'], immediately: true },
+  { id: '@deepseek-ai/dsh-client-ui-theme', dir: 'ui-theme', url: '/plugins/ui-theme.js', rev: 'fx', inject: [], immediately: true },
+  { id: '@deepseek-ai/dsh-client-locale', dir: 'locale', url: '/plugins/locale.js', rev: 'fx', inject: [], immediately: true },
+  { id: '@deepseek-ai/dsh-client-ui-layout', dir: 'ui-layout', url: '/plugins/ui-layout.js', rev: 'fx', inject: ['@deepseek-ai/dsh-client-runtime'] },
+  { id: '@deepseek-ai/dsh-client-ui-sidebar', dir: 'ui-sidebar', url: '/plugins/ui-sidebar.js', rev: 'fx', inject: ['@deepseek-ai/dsh-client-ui-layout'] },
+  { id: '@deepseek-ai/dsh-client-ui-conversation', dir: 'ui-conversation', url: '/plugins/ui-conversation.js', rev: 'fx', inject: ['@deepseek-ai/dsh-client-ui-layout'] },
+  {
+    id: '@deepseek-ai/dsh-client-ui-workspace',
+    dir: 'ui-workspace',
+    url: '/plugins/ui-workspace.js',
+    rev: 'fx',
+    inject: [
+      '@deepseek-ai/dsh-client-runtime',
+      '@deepseek-ai/dsh-client-ui-conversation',
+      '@deepseek-ai/dsh-client-ui-sidebar',
+    ],
+  },
+  { id: '@deepseek-ai/dsh-client-ui-trajectory', dir: 'ui-trajectory', url: '/plugins/ui-trajectory.js', rev: 'fx', inject: ['@deepseek-ai/dsh-client-ui-conversation'] },
+]
+
+const bundles = new Map(PLUGINS.map(plugin => [
+  plugin.url,
+  readFileSync(join(process.cwd(), 'packages/client', plugin.dir, 'lib/client.js'), 'utf8'),
+]))
+
+interface FixtureWindow extends Window {
+  __DSH_BOOT__?: { rev: string; entries: WebBootEntry[] }
+  __ModuleLoader__?: unknown
+}
+
+class ResizeObserverStub {
+  observe(): void {}
+  disconnect(): void {}
+  unobserve(): void {}
+}
+
+const win = window as FixtureWindow
+let unmount: (() => void) | undefined
+
+/** Normalize a rendered search card to a stable text shape: the kind, the banner
+ *  summary, each file header (path + count), each visible match line, the expand
+ *  control label, and the recovery footer. CSS-module class names carry a
+ *  per-build hash in one of two schemes — ui-primitives emits `_<name>_<hash>`
+ *  (name bounded by underscores), ui-conversation emits `<hash>_<name>` (name at
+ *  the end). `hasClass` matches a module class by its logical name under either,
+ *  without matching a longer name that contains it (`line` must not hit
+ *  `lineNumber`). */
+function hasClass(el: Element, name: string): boolean {
+  return [...el.classList].some(cls => cls === name || cls.endsWith(`_${name}`) || cls.startsWith(`_${name}_`) || cls.includes(`_${name}_`))
+}
+
+function cardShape(root: Element): string {
+  const card = root.querySelector('[data-search]')
+  if (card === null) return '<no search card>'
+  const pick = (from: Element, name: string): Element[] =>
+    [...from.querySelectorAll('*')].filter(el => hasClass(el, name))
+  const lines: string[] = [`kind=${card.getAttribute('data-search')}`]
+  const summary = pick(card, 'summary')[0]?.textContent?.trim()
+  if (summary !== undefined && summary !== '') lines.push(`summary=${summary}`)
+  for (const header of pick(card, 'fileHeader')) lines.push(`file=${header.textContent?.trim() ?? ''}`)
+  for (const row of pick(card, 'line')) lines.push(`line=${row.textContent?.trim() ?? ''}`)
+  const expand = pick(card, 'expand')[0]?.textContent?.trim()
+  if (expand !== undefined && expand !== '') lines.push(`expand=${expand}`)
+  const recovery = pick(root, 'searchRecovery')[0]?.textContent?.trim()
+  if (recovery !== undefined && recovery !== '') lines.push(`recovery=${recovery}`)
+  return lines.join('\n')
+}
+
+beforeEach(() => {
+  localStorage.clear()
+  // English pinned before boot so the sidebar's role/text locators stay
+  // deterministic (the built-boot smoke's convention).
+  localStorage.setItem('dsh.locale', 'en')
+  document.title = 'DeepSeek Harness'
+  vi.stubGlobal('ResizeObserver', ResizeObserverStub)
+  vi.stubGlobal('requestAnimationFrame', (callback: FrameRequestCallback) =>
+    setTimeout(() => { callback(0) }, 0) as unknown as number)
+  vi.stubGlobal('cancelAnimationFrame', (id: number) => { clearTimeout(id) })
+})
+
+afterEach(() => {
+  act(() => { unmount?.() })
+  unmount = undefined
+  cleanup()
+  delete win.__DSH_BOOT__
+  delete win.__ModuleLoader__
+  document.body.innerHTML = ''
+  document.head.querySelectorAll('style[data-plugin]').forEach((style) => { style.remove() })
+  document.title = ''
+  history.replaceState(null, '', '/')
+  vi.unstubAllGlobals()
+})
+
+describe('assembled search card', () => {
+  it('renders the grep card, its truncation summary, and its capped head/tail slice from the built bundles', async () => {
+    history.replaceState(null, '', '/?fixture')
+    const root = document.createElement('div')
+    root.id = 'root'
+    document.body.appendChild(root)
+    win.__DSH_BOOT__ = { rev: 'fx', entries: PLUGINS.map(({ dir: _dir, ...plugin }) => plugin) }
+    act(() => {
+      const entry = new AppWebEntry(root, {
+        fetchBundle: (url) => {
+          const code = bundles.get(url)
+          return code === undefined ? Promise.reject(new Error(`missing built bundle ${url}`)) : Promise.resolve(code)
+        },
+        executeBundle: (code) => { (0, eval)(code) },
+      })
+      void entry.run()
+      unmount = () => { entry.dispose() }
+    })
+
+    const tree = await screen.findByRole('tree', { name: 'Sessions' }, { timeout: 10_000 })
+    fireEvent.click(await within(tree).findByText('Fixture 历史会话'))
+    // Wait for chat content to reach the fixture's later turns (the bash sample
+    // is turn 65, the grep card turn 66).
+    await waitFor(() => {
+      expect(document.querySelector('[data-sample="bash-global"]')).not.toBeNull()
+    }, { timeout: 10_000 })
+    // The grep turn's keyed SearchRow renders the card resident: wait for it.
+    await waitFor(() => {
+      const tools = [...document.querySelectorAll('[data-tool]')].map(el => el.getAttribute('data-tool'))
+      expect(tools, `tools present: ${tools.join(', ')}`).toContain('grep')
+    }, { timeout: 10_000 })
+
+    // `data-tool` sits on the summary row; the card and recovery footer are its
+    // siblings inside the SearchRow wrapper, so shape the wrapper (its parent).
+    const grepRow = document.querySelector('[data-tool="grep"]')!.parentElement!
+    const shape = cardShape(grepRow)
+    if (refreshing) {
+      mkdirSync(dirname(EXPECTED), { recursive: true })
+      writeFileSync(EXPECTED, shape)
+    }
+    await expect(shape).toMatchFileSnapshot(EXPECTED)
+  })
+})

apps/web/tests/seeded-history.e2e.ts

@@ -1,11 +1,12 @@
 // Web e2e scenario: seeded history. A recorded session seeded cold through
 // the REAL persistence API renders purely from the log — the surface nothing
 // else covers: sidebar cold listing, the implicit resume/attach inside the
-// history RPC, history-page tool views, and the client fold of historical
+// history RPC, history-page tool views, and the client's log-ordered transcript
 // events — with ZERO model calls in replay (no replay fixture; a stray stream
 // fails loud on the open llm seam). The cold session also carries the one
 // keyless command-row surface: an Access-chip pick runs `/permission` on the
-// host, so the settled row's copy has a golden here. The seed is a recorded fixture under the
+// host, so the settled row's copy has a golden here. The seed is a recorded
+// fixture under the
 // same record discipline as every other: DSH_SNAPSHOT=record drives the turn
 // live through the composer (real read tool against seeded workspace files)
 // and harvests seed.jsonl; replay/refresh seed it cold and only render.
@@ -34,6 +35,90 @@ const SEED_ID = 'seeded-history-web-e2e'
 
 const PROMPT = 'Use the read tool twice in one assistant message: read a.txt and b.txt. Then reply with the single word DONE and stop.'
 
+/**
+ * Append a complete, valid compaction transaction over the recorded turn's own
+ * surface. The recording stays model-authentic and reusable; replay adds this
+ * deterministic condition before seeding it cold, so the scenario pins the bug
+ * this change fixes — a landed compaction must not erase history the reader
+ * already saw — through the real host and the real browser.
+ * @param raw - the committed seed fixture text.
+ * @returns the fixture with a compacted turn appended.
+ */
+function withCompaction(raw: string): string {
+  const lines = raw.trimEnd().split('\n')
+  const events = lines.slice(1).map(line => JSON.parse(line) as {
+    type: string
+    seq: number
+    time: number
+    surfaceOp?: unknown
+    data?: { turn?: unknown }
+  })
+  const surfaceSeqs = events
+    .filter(event => event.surfaceOp === 'append'
+      && (event.type === 'user/message'
+        || event.type === 'assistant/message'
+        || event.type === 'tool/result'
+        || event.type === 'steering/message'))
+    .map(event => event.seq)
+  const first = surfaceSeqs[0]
+  const last = surfaceSeqs.at(-1)
+  const tail = events.at(-1)
+  if (first === undefined || last === undefined || tail === undefined) {
+    throw new Error('seeded-history compaction requires a non-empty closed surface')
+  }
+  // The transaction opens the turn after the recording's last closed one; read
+  // it from the fixture so a re-recording with a different turn count stays
+  // valid instead of appending a duplicate turn number.
+  const lastTurn = events.filter(event => event.type === 'turn/end').at(-1)?.data?.turn
+  if (typeof lastTurn !== 'number') {
+    throw new Error('seeded-history compaction requires a recording ending on a closed turn')
+  }
+  const turn = lastTurn + 1
+  let seq = tail.seq + 1
+  let time = tail.time + 1
+  /**
+   * Append one event at the next seq/time.
+   * @param event - the event body, without seq/time.
+   * @returns the seq it took, so provenance cites the push instead of arithmetic over the push order below.
+   */
+  const at = (event: Record<string, unknown>): number => {
+    const taken = seq++
+    lines.push(JSON.stringify({ ...event, seq: taken, time: time++ }))
+    return taken
+  }
+  at({ type: 'turn/start', data: { turn, trigger: { kind: 'injection', source: { kind: 'plugin', plugin: 'compact' } } } })
+  const startSeq = at({ type: 'compact/start', data: { turn } })
+  const summarySeq = at({
+    type: 'compact/summary',
+    data: {
+      summary: [{
+        type: 'text',
+        text: '## Cold resume compact summary\n\n- The exact summary remains available.',
+      }],
+      shadowedRange: { start: first, end: last },
+      shadowedSeqs: surfaceSeqs,
+      shadowedTokenCount: 10_000,
+      provider: 'snapshot',
+      model: 'snapshot-compactor',
+    },
+  })
+  at({
+    type: 'user/message',
+    data: {
+      content: [{
+        type: 'text',
+        text: '<context_checkpoint>Model-only compact checkpoint.</context_checkpoint>',
+      }],
+      source: { kind: 'plugin', plugin: 'compact' },
+    },
+    surfaceOp: { op: 'replace', start: first, end: last },
+    sourceEventSeqs: [startSeq, summarySeq, ...surfaceSeqs],
+  })
+  at({ type: 'compact/end', data: { turn } })
+  at({ type: 'turn/end', data: { turn, reason: { kind: 'completed' } } })
+  return `${lines.join('\n')}\n`
+}
+
 describe('web e2e: seeded history renders through cold resume', () => {
   let scaffold: WebScaffold
   let browser: Browser
@@ -53,7 +138,7 @@ describe('web e2e: seeded history renders through cold resume', () => {
     if (MODE !== 'record') {
       const raw = await readFile(SEED, 'utf8')
       expect(fixtureUserPrompts(raw), 'seed fixture must carry exactly the drive prompt').toEqual([PROMPT])
-      await seedSession(scaffold, raw, SEED_ID)
+      await seedSession(scaffold, withCompaction(raw), SEED_ID)
     }
     browser = await chromium.launch()
     page = await newEnglishPage(browser)
@@ -119,11 +204,15 @@ describe('web e2e: seeded history renders through cold resume', () => {
     await sessionRow.click()
     // Settled barrier for history: the recorded final assistant text renders.
     await expect.poll(() => page.getByText('DONE', { exact: true }).count(), { timeout: 15_000 }).toBe(1)
+    await expect.poll(() => page.getByText('Context compacted', { exact: true }).count(), { timeout: 10_000 }).toBe(1)
     // Tool cards render from logged tool/call + tool/result alone (views are
     // host-recomputed per page; the generic card is the documented default).
     const toolRows = page.locator('[data-variant], [data-sample]')
     await expect.poll(() => toolRows.count(), { timeout: 10_000 }).toBeGreaterThanOrEqual(2)
     expect(await page.getByText('a.txt', { exact: false }).count()).toBeGreaterThan(0)
+    // The bug this fixes: the compaction shadowed the whole recorded surface on
+    // the model side, and the prompt and full tool output are still on screen.
+    expect(await page.getByText(PROMPT, { exact: true }).count()).toBe(1)
 
     const agent = scaffold.ctx.agents.get(SessionId(SEED_ID))
     if (agent === undefined) throw new Error('seeded session did not attach an agent')
@@ -230,6 +319,22 @@ describe('web e2e: seeded history renders through cold resume', () => {
     await expect.poll(() => page.getByText('a.txt', { exact: false }).count(), { timeout: 5_000 }).toBeGreaterThan(0)
   })
 
+  it.skipIf(MODE === 'record')('expands the cold-resumed compact summary', async () => {
+    onTestFailed(() => saveFailureShot(page, 'web-e2e-seeded-compaction'))
+    const marker = page.getByRole('button', { name: /Context compacted/ })
+    await marker.waitFor({ timeout: 10_000 })
+    expect(await marker.getAttribute('aria-expanded')).toBe('false')
+    await marker.click()
+    await expect.poll(() => marker.getAttribute('aria-expanded'), { timeout: 5_000 }).toBe('true')
+    await expect.poll(() => page.getByRole('heading', { name: 'Cold resume compact summary' }).count(), {
+      timeout: 5_000,
+    }).toBe(1)
+    expect(await page.getByText('The exact summary remains available.', { exact: false }).count()).toBeGreaterThan(0)
+    // Restore the shared page state for any later case.
+    await marker.click()
+    await expect.poll(() => marker.getAttribute('aria-expanded'), { timeout: 5_000 }).toBe('false')
+  })
+
   it.skipIf(MODE === 'record')('an Access-chip switch lands one command row: bare name, non-repeating settlement text', async () => {
     onTestFailed(() => saveFailureShot(page, 'web-e2e-seeded-command-row'))
     // The Access chip submits `/permission <preset>` — a host command with no

apps/web/tests/settings-chrome.e2e.ts

@@ -18,7 +18,7 @@ import {
   acknowledgeReloadConnectionLoss, assertFixtureInventory, captureStableAria, compareOrRefreshGolden,
   launchWebScaffold, watchConsole, webSnapshotMode, type WebScaffold,
 } from './scaffold.ts'
-import { saveFailureShot } from './support.ts'
+import { ZH_BROWSER_LOCALE, saveFailureShot } from './support.ts'
 
 const SNAPSHOT_DIR = fileURLToPath(new URL('./snapshots/settings-chrome', import.meta.url))
 const DIALOG_EXPECTED = join(SNAPSHOT_DIR, 'dialog.expected.md')
@@ -33,7 +33,9 @@ describe('web e2e: settings modal and General preferences', () => {
   beforeAll(async () => {
     scaffold = await launchWebScaffold({})
     browser = await chromium.launch()
-    page = await browser.newPage({ viewport: { width: 1680, height: 1000 } })
+    // Chinese browser: the shared page asserts the localized settings surface
+    // the client derives from it (the English default has its own spec below).
+    page = await browser.newPage({ viewport: { width: 1680, height: 1000 }, locale: ZH_BROWSER_LOCALE })
     tripwire = watchConsole(page)
     await page.goto(scaffold.baseUrl, { waitUntil: 'load' })
     await page.waitForSelector('[class*="frame"]', { timeout: 30_000 })
@@ -215,6 +217,30 @@ describe('web e2e: settings modal and General preferences', () => {
     expect(tripwire.pageErrors).toEqual([])
   }, 90_000)
 
+  it('opens an English browser in English without any stored preference', async () => {
+    // A second page under a different browser language: nothing is persisted
+    // for it, so the settings surface must follow the browser rather than the
+    // product fallback the shared zh page shows.
+    const enPage = await browser.newPage({ viewport: { width: 1680, height: 1000 }, locale: 'en-US' })
+    const enTripwire = watchConsole(enPage)
+    onTestFailed(() => saveFailureShot(enPage, 'web-e2e-settings-browser-language'))
+    try {
+      await enPage.goto(scaffold.baseUrl, { waitUntil: 'load' })
+      await enPage.waitForSelector('[class*="frame"]', { timeout: 30_000 })
+      expect(await enPage.evaluate(() => localStorage.getItem('dsh.locale'))).toBeNull()
+      await enPage.getByRole('button', { name: 'Settings', exact: true }).click()
+      const dialog = enPage.getByRole('dialog', { name: 'Settings' })
+      await dialog.waitFor({ timeout: 10_000 })
+      await dialog.getByRole('button', { name: 'English' }).waitFor({ timeout: 10_000 })
+      // This page has no closing inventory spec to sweep its console, so the
+      // scenario clears both tripwire channels itself.
+      expect(enTripwire.pageErrors).toEqual([])
+      expect(enTripwire.warnings).toEqual([])
+    } finally {
+      await enPage.close()
+    }
+  }, 90_000)
+
   it.skipIf(MODE === 'record')('keeps the fixture inventory closed', async () => {
     expect(tripwire.warnings).toEqual([])
     await assertFixtureInventory(SNAPSHOT_DIR, ['dialog.expected.md'])

apps/web/tests/smoke-real.e2e.ts

@@ -163,6 +163,8 @@ describe('dsh web keyless CLI smoke', () => {
         env: {
           ...process.env,
           DEEPSEEK_API_KEY: 'keyless-web-no-call',
+          DSH_HOME: join(sessionsDir, '.dsh'),
+          DSH_AGENTS_HOME: join(sessionsDir, '.agents'),
           TSX_TSCONFIG_PATH: join(REPO_ROOT, 'tsconfig.json'),
         },
         stdio: ['ignore', 'pipe', 'pipe'],
@@ -226,6 +228,7 @@ describe('dsh web keyless CLI smoke', () => {
           DEEPSEEK_API_KEY: 'keyless-web-workspace',
           DEEPSEEK_BASE_URL: `http://127.0.0.1:${address.port}`,
           DSH_HOME: join(workspace, '.dsh'),
+          DSH_AGENTS_HOME: join(workspace, '.agents'),
           TSX_TSCONFIG_PATH: join(REPO_ROOT, 'tsconfig.json'),
         },
         stdio: ['ignore', 'pipe', 'pipe'],
@@ -245,6 +248,8 @@ describe('dsh web keyless CLI smoke', () => {
           setTimeout(() => { reject(new Error('provider request not received in 10s')) }, 10_000).unref()
         }),
       ])
+      expect(captured.messages?.some(message =>
+        message.role === 'user' && message.content?.includes('<available_skills>'))).toBe(false)
       const workspaceMessage = captured.messages?.find(message =>
         message.role === 'user' && message.content?.includes('web-workspace-context-probe'))
       expect(workspaceMessage).toMatchInlineSnapshot(`
@@ -413,6 +418,7 @@ describe('dsh web keyless CLI smoke', () => {
           DEEPSEEK_BASE_URL: `http://127.0.0.1:${address.port}`,
           DSH_TOOLS_MODE: 'code',
           DSH_HOME: join(workspace, '.dsh'),
+          DSH_AGENTS_HOME: join(workspace, '.agents'),
           TSX_TSCONFIG_PATH: join(REPO_ROOT, 'tsconfig.json'),
         },
         stdio: ['ignore', 'pipe', 'pipe'],
@@ -461,8 +467,8 @@ describe.skipIf(!process.env.DEEPSEEK_API_KEY || notReady.length > 0)('web smoke
     sessionsDir = mkdtempSync(join(tmpdir(), 'dsh-web-w5-'))
     const port = await probeFreePort()
     // tsx boot mirrors demo:web — lib/ may be unbuilt in this worktree. Isolate
-    // the global Harness home inside the temp world; tsx also needs the repo's
-    // loader and tsconfig paths pointed at explicitly.
+    // the host-level Harness and shared-agent homes inside the temp world; tsx
+    // also needs the repo's loader and tsconfig paths pointed at explicitly.
     const tsxLoader = pathToFileURL(createRequire(join(REPO_ROOT, 'package.json')).resolve('tsx')).href
     child = spawn(
       process.execPath,
@@ -472,6 +478,7 @@ describe.skipIf(!process.env.DEEPSEEK_API_KEY || notReady.length > 0)('web smoke
         env: {
           ...process.env,
           DSH_HOME: join(sessionsDir, '.dsh'),
+          DSH_AGENTS_HOME: join(sessionsDir, '.agents'),
           TSX_TSCONFIG_PATH: join(REPO_ROOT, 'tsconfig.json'),
         },
         stdio: ['ignore', 'pipe', 'pipe'],

apps/web/tests/snapshots/code-mode-round/ui.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - 'button "Think The user wants me to write a single `run_code` program that:"':
   - img
   - img
@@ -20,10 +22,8 @@
   - img
   - text: Code Run bash echo and catch missing file read
 - img
-- text: Bash Echo CODE_ROUND_OK
-- 'button "Read Error: cannot read \"{{cwd}}/workspace/missing.txt\": not found"':
-  - img
-  - text: "Read Error: cannot read \"{{cwd}}/workspace/missing.txt\": not found"
+- text: Bash Echo CODE_ROUND_OK 失败 Read
+- button "missing.txt"
 - button "Think The program ran successfully. Let me now reply DONE as instructed.":
   - img
   - img
@@ -42,4 +42,4 @@
   - text: DeepSeek-V4-Flash
   - img
 - button "Send message" [disabled]
-- text: 1 turns · 2 steps Tool call {{duration}} Cache hit 52% Input 17.2K tok · Output 252 tok
+- text: 1 turns · 2 steps Tool call {{duration}} Context 7% of 128K Cache hit 52% Input 17.2K tok · Output 252 tok

apps/web/tests/snapshots/cordis-tool-round/ui.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - button "Think The user wants me to:":
   - img
   - img
@@ -57,4 +59,4 @@
   - text: DeepSeek-V4-Flash
   - img
 - button "Send message" [disabled]
-- text: 1 turns · 4 steps Tool call {{duration}} Cache hit 77% Input 66.5K tok · Output 312 tok
+- text: 1 turns · 4 steps Tool call {{duration}} Context 13% of 128K Cache hit 77% Input 66.5K tok · Output 312 tok

apps/web/tests/snapshots/fresh-round-trip/ui.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - button "Think The user wants me to run a simple bash command and reply with \"DONE\".":
   - img
   - img
@@ -37,4 +39,4 @@
   - text: DeepSeek-V4-Flash
   - img
 - button "Send message" [disabled]
-- text: 1 turns · 2 steps Tool call {{duration}} Cache hit 99% Input 15.7K tok · Output 111 tok
+- text: 1 turns · 2 steps Tool call {{duration}} Context 6% of 128K Cache hit 99% Input 15.7K tok · Output 111 tok

apps/web/tests/snapshots/lifecycle-chrome/reloaded.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - button "Think The user wants me to reply with a single word. Let me comply.":
   - img
   - img
@@ -29,4 +31,4 @@
   - text: DeepSeek-V4-Flash
   - img
 - button "Send message" [disabled]
-- text: 1 turns · 1 steps Cache hit 99% Input 7.8K tok · Output 21 tok
+- text: 1 turns · 1 steps Context 6% of 128K Cache hit 99% Input 7.8K tok · Output 21 tok

apps/web/tests/snapshots/live-interactions/cancel.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - paragraph: partial
 - text: Stopped
 - button "Copy":

apps/web/tests/snapshots/live-interactions/error-auth.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - textbox "Message the agent"
 - button "Commands":
   - img

apps/web/tests/snapshots/live-interactions/loading.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - paragraph: partial
 - status: Deep diving...
 - textbox "Message the agent"

apps/web/tests/snapshots/live-interactions/retry.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - group:
   - status: Retried model request (1/2) · {{duration}}
 - button "Think The user is asking for a one-sentence description of event sourcing. This is a straightforward knowledge question that doesn't require any skill loading or tool calls.":
@@ -31,4 +33,4 @@
   - text: DeepSeek-V4-Flash
   - img
 - button "Send message" [disabled]
-- text: 1 turns · 1 steps Cache hit 99% Input 7.8K tok · Output 79 tok
+- text: 1 turns · 1 steps Context 6% of 128K Cache hit 99% Input 7.8K tok · Output 79 tok

apps/web/tests/snapshots/message-actions/ui.expected.md

@@ -10,22 +10,16 @@
 - tooltip "Copy"
 - button "Branch into a new conversation":
   - img
-- button "Edit":
-  - img
 - button "Think The user wants me to read a.txt and b.txt, then reply with \"DONE\". Let me do both reads in parallel.":
   - img
   - img
   - text: Think The user wants me to read a.txt and b.txt, then reply with "DONE". Let me do both reads in parallel.
-- button "Read a.txt":
-  - img
-  - img
-  - text: Read
-  - button "a.txt"
-- button "Read b.txt":
-  - img
-  - img
-  - text: Read
-  - button "b.txt"
+- img
+- text: Read
+- button "a.txt"
+- img
+- text: Read
+- button "b.txt"
 - button "Think Both files have been read. a.txt contains \"alpha\" and b.txt contains \"beta\". I'll now reply with DONE as instructed.":
   - img
   - img

apps/web/tests/snapshots/onboarding-deepseek-config/welcome.expected.md

@@ -1,10 +1,9 @@
 - region "内测声明":
   - heading "内测声明" [level=2]
-  - paragraph: 感谢您愿意拨冗试用 DeepSeek Harness。
-  - paragraph: 目前的版本仍处于内部测试阶段，功能仍待完善，体验难免有些粗糙。
+  - paragraph: 感谢您愿意拨冗试用 DeepSeek Harness。当前版本仍处于内部测试阶段，功能仍待完善，体验难免有些粗糙。
   - blockquote: “如切如磋，如琢如磨。” 产品的成长，离不开一次次真实的碰撞与坦诚的反馈。您在真实使用中发现的问题，也可能促使我们重新审视，甚至推翻已有的设计。
   - paragraph:
-    - text: 我们尤其希望听见那些失败、困惑与不顺手的时刻——
+    - text: 为了帮助我们更准确地还原您真实使用中的问题，内测版本默认会上传所有 Session Log；如需关闭，可以设置环境变量 DSH_TELEMETRY_DISABLED=1。另外，
     - strong: 如果您有任何反馈与建议，请在企业微信群中留言告诉我们
     - text: 。每一条反馈，都会帮助我们把它打磨得更好。
   - button "继续"

apps/web/tests/snapshots/plan-review/approved.expected.md

@@ -10,8 +10,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - 'button "Think The user wants me to plan a small change to add a `--greeting` flag to a CLI. They explicitly told me not to read or write any files, and to call exit_plan_mode with a short plan. Let me do that directly."':
   - img
   - img
@@ -42,4 +44,4 @@
   - text: DeepSeek-V4-Flash
   - img
 - button "Send message" [disabled]
-- text: 1 turns · 2 steps Tool call {{duration}} Cache hit 51% Input 10.2K tok · Output 346 tok
+- text: 1 turns · 2 steps Tool call {{duration}} Context 4% of 128K Cache hit 51% Input 10.2K tok · Output 346 tok

apps/web/tests/snapshots/question-composer/answered.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - button "Think The user wants me to use the ask_user_question tool with specific parameters. Let me do exactly that.":
   - img
   - img
@@ -37,4 +39,4 @@
   - text: DeepSeek-V4-Flash
   - img
 - button "Send message" [disabled]
-- text: 1 turns · 2 steps Tool call {{duration}} Cache hit 95% Input 8.6K tok · Output 180 tok
+- text: 1 turns · 2 steps Tool call {{duration}} Context 3% of 128K Cache hit 95% Input 8.6K tok · Output 180 tok

apps/web/tests/snapshots/queue-actions/collapsed.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - paragraph: partial
 - status: Deep diving...
 - button "2 queued messages"

apps/web/tests/snapshots/queue-actions/editing.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - paragraph: partial
 - status: Deep diving...
 - button "2 queued messages" [disabled] [expanded]

apps/web/tests/snapshots/queue-actions/ui.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - paragraph: partial
 - status: Deep diving...
 - list:

apps/web/tests/snapshots/search-card/grep-card.expected.txt

@@ -0,0 +1,11 @@
+kind=matches
+summary=显示 9 / 共 42 处匹配 · 3 个文件
+file=packages/client/ui-primitives/src/SearchBlock.tsx3
+file=packages/client/ui-conversation/src/client/toolviews/search-row.tsx4
+line=16: export const DEFAULT_SEARCH_MAX_LINES = 16
+line=138: export function SearchBlock(props: SearchBlockProps) {
+line=141:   const [collapsed, setCollapsed] = useState<ReadonlySet<number>>(() => new Set())
+line=73:   const search = searchCardModel(block)
+line=90:         <SearchBlock {...search.card} maxLines={CHAT_SEARCH_MAX_LINES} className={css.search} />
+line=113:     ctx.slots.register({ name: 'conversation.chat.toolview', key: 'grep' }, SearchRow)
+expand=… 其余 4 行

apps/web/tests/snapshots/seeded-history/command-row.expected.md

@@ -9,22 +9,16 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
-  - img
 - button "Think The user wants me to read a.txt and b.txt, then reply with \"DONE\". Let me do both reads in parallel.":
   - img
   - img
   - text: Think The user wants me to read a.txt and b.txt, then reply with "DONE". Let me do both reads in parallel.
-- button "Read a.txt":
-  - img
-  - img
-  - text: Read
-  - button "a.txt"
-- button "Read b.txt":
-  - img
-  - img
-  - text: Read
-  - button "b.txt"
+- img
+- text: Read
+- button "a.txt"
+- img
+- text: Read
+- button "b.txt"
 - button "Think Both files have been read. a.txt contains \"alpha\" and b.txt contains \"beta\". I'll now reply with DONE as instructed.":
   - img
   - img
@@ -35,6 +29,9 @@
 - button "Branch into a new conversation":
   - img
 - text: 7/25 {{clock}}
+- button "Context compacted View compaction summary":
+  - img
+  - text: Context compacted View compaction summary
 - button "Context injection":
   - img
   - img

apps/web/tests/snapshots/seeded-history/ui.expected.md

@@ -9,22 +9,16 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
-  - img
 - button "Think The user wants me to read a.txt and b.txt, then reply with \"DONE\". Let me do both reads in parallel.":
   - img
   - img
   - text: Think The user wants me to read a.txt and b.txt, then reply with "DONE". Let me do both reads in parallel.
-- button "Read a.txt":
-  - img
-  - img
-  - text: Read
-  - button "a.txt"
-- button "Read b.txt":
-  - img
-  - img
-  - text: Read
-  - button "b.txt"
+- img
+- text: Read
+- button "a.txt"
+- img
+- text: Read
+- button "b.txt"
 - button "Think Both files have been read. a.txt contains \"alpha\" and b.txt contains \"beta\". I'll now reply with DONE as instructed.":
   - img
   - img
@@ -35,6 +29,9 @@
 - button "Branch into a new conversation":
   - img
 - text: 7/25 {{clock}}
+- button "Context compacted View compaction summary":
+  - img
+  - text: Context compacted View compaction summary
 - button "Context injection":
   - img
   - img

apps/web/tests/snapshots/steering/mid-steer.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - button "Think The user wants me to use the ask_user_question tool to ask them a specific question with the given parameters. Let me do exactly that.":
   - img
   - img

apps/web/tests/snapshots/steering/settled.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - button "Think The user wants me to use the ask_user_question tool to ask them a specific question with the given parameters. Let me do exactly that.":
   - img
   - img
@@ -38,4 +40,4 @@
   - text: DeepSeek-V4-Flash
   - img
 - button "Send message" [disabled]
-- text: 1 turns · 2 steps Tool call {{duration}} Cache hit 98% Input 15.8K tok · Output 156 tok
+- text: 1 turns · 2 steps Tool call {{duration}} Context 6% of 128K Cache hit 98% Input 15.8K tok · Output 156 tok

apps/web/tests/snapshots/web-search-round/ui.expected.md

@@ -9,8 +9,10 @@
   - img
 - button "Branch into a new conversation":
   - img
-- button "Edit":
+- button "Context injection":
   - img
+  - img
+  - text: Context injection
 - img
 - text: Search DeepSeek Harness snapshot search
 - list:
@@ -32,4 +34,4 @@
   - text: DeepSeek-V4-Flash
   - img
 - button "Send message" [disabled]
-- text: 1 turns · 2 steps Tool call {{duration}} Cache hit 0% Input 22 tok · Output 7 tok
+- text: 1 turns · 2 steps Tool call {{duration}} Context 0% of 128K Cache hit 0% Input 22 tok · Output 7 tok

apps/web/tests/support.ts

@@ -9,11 +9,18 @@ export const DIST_INDEX = fileURLToPath(new URL('../dist/index.html', import.met
 
 export const REPO_ROOT = fileURLToPath(new URL('../../..', import.meta.url))
 
+/**
+ * Browser language a page must advertise to boot into the product's Chinese
+ * surface: with no stored preference the client derives its initial locale
+ * from the browser, and Playwright's default browser asks for English.
+ */
+export const ZH_BROWSER_LOCALE = 'zh-CN'
+
 /**
  * Open the standard browser-test page with English selected before client
  * boot. This keeps role locators and goldens deterministic across localized
- * component migrations; the settings locale scenario deliberately bypasses
- * this helper to cover the product's default Chinese state.
+ * component migrations; the scenarios asserting the Chinese surface bypass
+ * this helper and advertise {@link ZH_BROWSER_LOCALE} instead.
  * @param browser - Playwright browser owning the page.
  * @param height - Viewport height; width is fixed to the lane baseline.
  * @returns the initialized page.

apps/web/tsconfig.json

@@ -46,6 +46,7 @@
     "tests/message-actions.e2e.ts",
     "tests/queue-actions.e2e.ts",
     "tests/skill-invocation-policy.e2e.ts",
+    "tests/permission-policy-context.e2e.ts",
     "tests/access-confirmation.e2e.ts"
   ],
   "references": [