docs: rebuild documentation site with VitePress (#518)

* docs: rebuild documentation site with VitePress * docs: address VitePress review feedback * docs: complete locale redirect validation
2026-11-11 08:38:48 +00:00 · 2026-09-04 14:50:50 +08:00 · 2026-09-04 14:50:50 +08:00 · 88ed21165b
commit 88ed21165b
parent 3f2eb6235f
51 changed files with 4495 additions and 1309 deletions
--- a/.github/workflows/_build-docs.yml
+++ b/.github/workflows/_build-docs.yml
@ -28,6 +28,7 @@ jobs:
    steps:
      - uses: actions/checkout@d23441a48e516b6c34aea4fa41551a30e30af803 # v6
        with:
+          fetch-depth: 0
          persist-credentials: false

      - name: Set up Node
--- a/.github/workflows/ci-docs.yml
+++ b/.github/workflows/ci-docs.yml
@ -11,11 +11,15 @@ on:
      - 'README_ZH.md'
      - 'docs/**'
      - 'github-pages/**'
+      - 'reme/config/default.yaml'
+      - 'integrations/claude_code/README.md'
+      - 'integrations/hermes_agent/README.md'
      - 'reme_studio/README*.md'
      - 'reme_studio/public/og.jpg'
      - 'typescript/README*.md'
      - 'plugins/*/README*.md'
      - 'benchmark/*/README*.md'
+      - 'benchmark/toolmemory/gitcha.png'
  pull_request:
    branches: [main, master, dev, develop]
    paths:
@ -26,11 +30,15 @@ on:
      - 'README_ZH.md'
      - 'docs/**'
      - 'github-pages/**'
+      - 'reme/config/default.yaml'
+      - 'integrations/claude_code/README.md'
+      - 'integrations/hermes_agent/README.md'
      - 'reme_studio/README*.md'
      - 'reme_studio/public/og.jpg'
      - 'typescript/README*.md'
      - 'plugins/*/README*.md'
      - 'benchmark/*/README*.md'
+      - 'benchmark/toolmemory/gitcha.png'
  workflow_dispatch:

 concurrency:
--- a/.github/workflows/deploy-docs.yml
+++ b/.github/workflows/deploy-docs.yml
@ -6,6 +6,9 @@ on:
    paths:
      - "github-pages/**"
      - "docs/**"
+      - "reme/config/default.yaml"
+      - "integrations/claude_code/README.md"
+      - "integrations/hermes_agent/README.md"
      - "README.md"
      - "README_ZH.md"
      - "reme_studio/README*.md"
@ -13,6 +16,7 @@ on:
      - "typescript/README*.md"
      - "plugins/*/README*.md"
      - "benchmark/*/README*.md"
+      - "benchmark/toolmemory/gitcha.png"
      - "AGENTS.md"
      - ".github/workflows/deploy-docs.yml"
      - ".github/workflows/_build-docs.yml"
--- a/AGENTS.md
+++ b/AGENTS.md
@ -62,6 +62,8 @@ and concise documentation together.
  conventions.
 - `benchmark/` and `cookbook/`: runnable evaluations and example workflows.
 - `docs/`: README-linked supporting pages and figures.
+- `github-pages/`: VitePress build shell, generated-content assembly, documentation checks, and GitHub Pages output. The
+  canonical theme and guides remain under `docs/`; `.generated/` and `dist/` are disposable.

 ## Development Setup

@ -195,6 +197,9 @@ Integration tests may contact real model providers, services, or agent subproces
 run credentialed or externally mutating tests automatically; run them only when the task requires them and the necessary
 environment has been supplied or authorized. Mock network, model, and subprocess boundaries in unit tests.

+If documentation or the documentation theme changes, run `npm test` and `npm run build` from `github-pages/`. The Job
+reference is generated from `reme/config/default.yaml`; do not edit generated pages directly.
+
 ## Change Guardrails

 - Preserve unrelated user changes in a dirty working tree.
--- a/README.md
+++ b/README.md
@ -51,12 +51,12 @@ users retain control of the durable files.

 - [2026.08] - Published [`@agentscope-ai/reme`](https://www.npmjs.com/package/@agentscope-ai/reme), providing native
  ReMe memory integrations for DeepSeek Harness and OpenClaw plus a shared TypeScript HTTP client.
- [2026.08] - Published the [ReMe blog](https://agentscope-ai.github.io/ReMe/?doc=en-reme-blog), an end-to-end introduction to its local-first memory
+- [2026.08] - Published the [ReMe blog](https://reme.agentscope.io/en/reme-blog), an end-to-end introduction to its local-first memory
  architecture, self-evolving workflows, hybrid search, proactive discovery, and benchmark results.
- [2026.08] - [Experience-driven enhancement method](https://reme.agentscope.io/?doc=toolmemory-en) of agent tool-use execution built
+- [2026.08] - [Experience-driven enhancement method](https://reme.agentscope.io/en/benchmarks/toolmemory) of agent tool-use execution built
  on ReMe is available on [arXiv:2608.03403](https://arxiv.org/abs/2608.03403).
- [2026.07] - Introduced optional plugins: [Daily Paper](https://reme.agentscope.io/?doc=daily-paper-en) for paper discovery and
-  analysis, and [Auto Fin](https://reme.agentscope.io/?doc=auto-fin-en) for researching the latest 24 hours of topic-related CLS news
+- [2026.07] - Introduced optional plugins: [Daily Paper](https://reme.agentscope.io/en/plugins/daily-paper) for paper discovery and
+  analysis, and [Auto Fin](https://reme.agentscope.io/en/plugins/auto-fin) for researching the latest 24 hours of topic-related CLS news
  with local-memory search and validated historical wikilinks.
 - [2026.07] - Our
  paper [Remember Me, Refine Me: A Dynamic Procedural Memory Framework for Experience-Driven Agent Evolution](https://aclanthology.org/2026.findings-acl.829/)
@ -171,7 +171,7 @@ Related: [[digest/wiki/memory-as-file.md]]

 The `core` installation includes Studio. After starting ReMe, open <http://127.0.0.1:2333/> to browse, edit, and search
 the workspace. To add Studio to a base installation, use `pip install "reme-ai[web]"`. See the
-[ReMe Studio guide](https://reme.agentscope.io/?doc=studio-en) for source builds, configuration, and development.
+[ReMe Studio guide](https://reme.agentscope.io/en/workspace/studio) for source builds, configuration, and development.

 ## 🤝 Use ReMe with Your Agent

@ -308,12 +308,12 @@ benchmark.

 | Benchmark                                                                   | Setting      |              Sample size | Agentic score | Focus                                                              |
 | --------------------------------------------------------------------------- | ------------ | -----------------------: | ------------: | ------------------------------------------------------------------ |
-| **[LongMemEval cleaned-s](https://reme.agentscope.io/?doc=longmemeval-en)** | **Overall**  |        **500 questions** |     **89.4%** | Cross-session retrieval, knowledge updates, and temporal reasoning |
-| [BEAM](https://reme.agentscope.io/?doc=beam-en)                             | 100K context | 20 cases / 400 questions |         66.1% | Ten types of long-context memory tasks                             |
-| [BEAM](https://reme.agentscope.io/?doc=beam-en)                             | 1M context   | 35 cases / 700 questions |         65.0% | Ultra-long conversation settings                                   |
+| **[LongMemEval cleaned-s](https://reme.agentscope.io/en/benchmarks/longmemeval)** | **Overall**  |        **500 questions** |     **89.4%** | Cross-session retrieval, knowledge updates, and temporal reasoning |
+| [BEAM](https://reme.agentscope.io/en/benchmarks/beam)                             | 100K context | 20 cases / 400 questions |         66.1% | Ten types of long-context memory tasks                             |
+| [BEAM](https://reme.agentscope.io/en/benchmarks/beam)                             | 1M context   | 35 cases / 700 questions |         65.0% | Ultra-long conversation settings                                   |

 ReMe also achieved a **0.580 PROC score across five user personas** in the repository's
-[π-Bench evaluation](https://reme.agentscope.io/?doc=pibench-en), 2.4% above NanoBot under the same test-model configuration. PROC
+[π-Bench evaluation](https://reme.agentscope.io/en/benchmarks/pibench), 2.4% above NanoBot under the same test-model configuration. PROC
 measures proactive handling of hidden intent, clarification, cross-session preferences and conventions, task
 dependencies, and underspecified requests.

@ -326,8 +326,8 @@ plugins; see the source distributions and their documentation for [Daily Paper](

 | Plugin                                                        | Capability                                                                                                    |
 | ------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------- |
-| [Daily Paper](https://reme.agentscope.io/?doc=daily-paper-en) | Discover and rank papers, analyze PDFs with an agent, and generate file-native notes and a five-minute brief. |
-| [Auto Fin](https://reme.agentscope.io/?doc=auto-fin-en)       | Fetch topic-related CLS news, search ReMe history, and generate wikilink-backed Markdown reports.             |
+| [Daily Paper](https://reme.agentscope.io/en/plugins/daily-paper) | Discover and rank papers, analyze PDFs with an agent, and generate file-native notes and a five-minute brief. |
+| [Auto Fin](https://reme.agentscope.io/en/plugins/auto-fin)       | Fetch topic-related CLS news, search ReMe history, and generate wikilink-backed Markdown reports.             |

 See [Plugin Management](docs/en/plugin_management.md) to install, inspect, validate, enable, and uninstall ReMe plugins.

@ -338,6 +338,8 @@ These guides cover the main user workflows and the runtime contracts implemented
 | Guide                                                                     | What you will learn                                                                                 |
 | ------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------- |
 | [Quick Start](docs/en/quick_start.md)                                     | Install ReMe, start the service, and run the first file and memory operations.                      |
+| [Configuration](docs/en/configuration.md)                                 | Configure the workspace, models, Service, Jobs, Components, plugins, and CLI overrides.             |
+| [Services and Deployment](docs/en/services.md)                            | Use HTTP, SSE, MCP, and Studio while respecting the default security boundary.                       |
 | [Memory as File](docs/en/memory_as_file.md)                               | Understand workspace layers, frontmatter, wikilinks, chunks, and the file-as-source-of-truth model. |
 | [Auto Memory](docs/en/auto_memory.md)                                     | Preserve source conversations and distill reusable daily memory cards.                              |
 | [Auto Resource](docs/en/auto_resource.md)                                 | Import supported text resources and turn them into source-linked daily cards.                       |
@ -347,7 +349,9 @@ These guides cover the main user workflows and the runtime contracts implemented
 | [Application Scenarios](docs/en/reme_scene.md)                            | Follow concrete financial research, coding-memory, and personal knowledge-base examples.           |
 | [Framework](docs/en/framework.md)                                         | Understand Application, Job, Step, Component, service, configuration, and lifecycle boundaries.     |
 | [TypeScript integrations](typescript/README.md)                            | Configure the shared client and native DeepSeek Harness and OpenClaw adapters.                       |
-| [ReMe Blog](https://agentscope-ai.github.io/ReMe/?doc=en-reme-blog)       | Read the product story, design rationale, examples, and benchmark summary.                          |
+| [CLI and Job API](docs/en/reference/cli.md)                               | Learn command syntax and use the generated default Job parameter reference.                         |
+| [Operations and Recovery](docs/en/operations.md)                          | Diagnose services, maintain indexes, and back up, migrate, or recover a workspace.                   |
+| [ReMe Blog](https://reme.agentscope.io/en/reme-blog)                      | Read the product story, design rationale, examples, and benchmark summary.                           |

 ## 🛠️ Common Commands

--- a/README_ZH.md
+++ b/README_ZH.md
@ -49,13 +49,13 @@

 - [2026.08] - 发布 [`@agentscope-ai/reme`](https://www.npmjs.com/package/@agentscope-ai/reme)，提供统一 TypeScript HTTP
  client，以及 DeepSeek Harness 和 OpenClaw 的原生 ReMe 记忆集成。
- [2026.08] - 发布 [ReMe 博客](https://agentscope-ai.github.io/ReMe/?doc=zh-reme-blog)，系统介绍本地优先的记忆架构、自进化工作流、混合检索、
+- [2026.08] - 发布 [ReMe 博客](https://reme.agentscope.io/zh/reme-blog)，系统介绍本地优先的记忆架构、自进化工作流、混合检索、
  主动发现与评测结果。
 - [2026.08] - 基于 ReMe 的智能体工具使用
-  [经验驱动增强方法](https://reme.agentscope.io/?doc=toolmemory-zh)已发布，见
+  [经验驱动增强方法](https://reme.agentscope.io/zh/benchmarks/toolmemory)已发布，见
  [arXiv:2608.03403](https://arxiv.org/abs/2608.03403)。
- [2026.07] - 新增可选插件：[每日论文](https://reme.agentscope.io/?doc=daily-paper-zh)用于论文发现与解析，
-  [Auto Fin](https://reme.agentscope.io/?doc=auto-fin-zh)用于研究最近 24 小时的主题相关财联社新闻，通过本地记忆搜索回顾历史材料并构建
+- [2026.07] - 新增可选插件：[每日论文](https://reme.agentscope.io/zh/plugins/daily-paper)用于论文发现与解析，
+  [Auto Fin](https://reme.agentscope.io/zh/plugins/auto-fin)用于研究最近 24 小时的主题相关财联社新闻，通过本地记忆搜索回顾历史材料并构建
  wikilink。
 - [2026.07] -
  我们的论文 [Remember Me, Refine Me: A Dynamic Procedural Memory Framework for Experience-Driven Agent Evolution](https://aclanthology.org/2026.findings-acl.829/)
@ -170,7 +170,7 @@ ReMe 会把 Agent 记忆保存为可读的 Markdown。

 上面的 `core` 安装已包含 Studio。启动 ReMe 后，打开 <http://127.0.0.1:2333/> 即可浏览、编辑和搜索 workspace。
 如需为基础安装单独添加 Studio，可使用 `pip install "reme-ai[web]"`。源码构建、配置和开发说明见
-[ReMe Studio 指南](https://reme.agentscope.io/?doc=studio-zh)。
+[ReMe Studio 指南](https://reme.agentscope.io/zh/workspace/studio)。

 ## 🤝 将 ReMe 接入你的 Agent

@ -301,11 +301,11 @@ ReMe 通过 Agent 多轮搜索与读取的方式，评测多会话和超长上

 | 基准                                                                        | 设置        |            样本量 | Agentic 得分 | 主要检验内容                   |
 | --------------------------------------------------------------------------- | ----------- | ----------------: | -----------: | ------------------------------ |
-| **[LongMemEval cleaned-s](https://reme.agentscope.io/?doc=longmemeval-zh)** | **整体**    |        **500 题** |    **89.4%** | 跨会话检索、知识更新与时间推理 |
-| [BEAM](https://reme.agentscope.io/?doc=beam-zh)                             | 100K 上下文 | 20 cases / 400 题 |        66.1% | 十类长上下文记忆任务           |
-| [BEAM](https://reme.agentscope.io/?doc=beam-zh)                             | 1M 上下文   | 35 cases / 700 题 |        65.0% | 超长对话设置                   |
+| **[LongMemEval cleaned-s](https://reme.agentscope.io/zh/benchmarks/longmemeval)** | **整体**    |        **500 题** |    **89.4%** | 跨会话检索、知识更新与时间推理 |
+| [BEAM](https://reme.agentscope.io/zh/benchmarks/beam)                             | 100K 上下文 | 20 cases / 400 题 |        66.1% | 十类长上下文记忆任务           |
+| [BEAM](https://reme.agentscope.io/zh/benchmarks/beam)                             | 1M 上下文   | 35 cases / 700 题 |        65.0% | 超长对话设置                   |

-在仓库的 [π-Bench 评测](https://reme.agentscope.io/?doc=pibench-zh)中，ReMe Agent 在 5 种用户角色上的平均 **PROC 得分为 0.580**
+在仓库的 [π-Bench 评测](https://reme.agentscope.io/zh/benchmarks/pibench)中，ReMe Agent 在 5 种用户角色上的平均 **PROC 得分为 0.580**
 ，比相同测试模型配置的 NanoBot 高 2.4%。PROC 用于评估隐藏意图完成、针对性澄清、跨会话偏好和规范复用、跨任务依赖推断以及欠规格请求推进等主动性能力。

 ## 🧩 扩展与插件
@ -316,8 +316,8 @@ ReMe 通过 Agent 多轮搜索与读取的方式，评测多会话和超长上

 | 插件                                                       | 能力                                                                           |
 | ---------------------------------------------------------- | ------------------------------------------------------------------------------ |
-| [每日论文](https://reme.agentscope.io/?doc=daily-paper-zh) | 发现并排序论文，使用 Agent 解读 PDF，生成文件化论文笔记和五分钟简报。          |
-| [Auto Fin](https://reme.agentscope.io/?doc=auto-fin-zh)    | 拉取主题相关财联社新闻，搜索 ReMe 历史材料并生成带 wikilink 的 Markdown 报告。 |
+| [每日论文](https://reme.agentscope.io/zh/plugins/daily-paper) | 发现并排序论文，使用 Agent 解读 PDF，生成文件化论文笔记和五分钟简报。          |
+| [Auto Fin](https://reme.agentscope.io/zh/plugins/auto-fin)    | 拉取主题相关财联社新闻，搜索 ReMe 历史材料并生成带 wikilink 的 Markdown 报告。 |

 安装、查看、校验、启用和卸载 ReMe 插件的方法见[插件管理](docs/zh/plugin_management.md)。

@ -328,6 +328,8 @@ ReMe 通过 Agent 多轮搜索与读取的方式，评测多会话和超长上
 | 文档                                                                     | 主要内容                                                               |
 | ------------------------------------------------------------------------ | ---------------------------------------------------------------------- |
 | [快速开始](docs/zh/quick_start.md)                                       | 安装 ReMe、启动服务，并执行首次文件和记忆操作。                        |
+| [基础配置](docs/zh/configuration.md)                                     | 配置 workspace、模型、Service、Job、Component、插件和命令行覆盖。      |
+| [服务与部署](docs/zh/services.md)                                        | 使用 HTTP、SSE、MCP 和 Studio，并理解默认安全边界。                    |
 | [Memory as File](docs/zh/memory_as_file.md)                              | 理解 workspace 分层、frontmatter、wikilink、chunk 和文件事实来源模型。 |
 | [Auto Memory](docs/zh/auto_memory.md)                                    | 保留过滤后的对话来源记录，并提炼可复用的 daily 记忆卡片。              |
 | [Auto Resource](docs/zh/auto_resource.md)                                | 导入支持的文本资料，转换为可追溯来源的 daily 卡片。                    |
@ -337,7 +339,9 @@ ReMe 通过 Agent 多轮搜索与读取的方式，评测多会话和超长上
 | [应用场景](docs/zh/reme_scene.md)                                        | 查看金融研究、研发记忆和个人知识库的完整使用示例。                     |
 | [框架说明](docs/zh/framework.md)                                         | 理解 Application、Job、Step、Component、service、配置和生命周期边界。  |
 | [TypeScript 集成](typescript/README_ZH.md)                               | 配置统一 client，以及 DeepSeek Harness 和 OpenClaw 原生适配器。        |
-| [ReMe 博客](https://agentscope-ai.github.io/ReMe/?doc=zh-reme-blog)      | 了解完整产品故事、设计动机、使用示例和评测摘要。                       |
+| [CLI 与 Job API](docs/zh/reference/cli.md)                               | 查询命令语法，以及由默认配置自动生成的 Job 参数参考。                  |
+| [运维与恢复](docs/zh/operations.md)                                      | 诊断服务、维护索引，并备份、迁移和恢复 workspace。                     |
+| [ReMe 博客](https://reme.agentscope.io/zh/reme-blog)                     | 了解完整产品故事、设计动机、使用示例和评测摘要。                       |

 ## 🛠️ 常用命令

--- a/benchmark/toolmemory/README.md
+++ b/benchmark/toolmemory/README.md
@ -6,7 +6,7 @@
 > Code: [https://github.com/WangCan1178/ExpG](https://github.com/WangCan1178/ExpG)

 <p align="center">
-  <img src="gitcha.png" alt="ExpG challenges and overview" width="85%">
+  <img src="./gitcha.png" alt="ExpG challenges and overview" width="85%">
 </p>

 ### Overview
--- a/benchmark/toolmemory/README_ZH.md
+++ b/benchmark/toolmemory/README_ZH.md
@ -6,7 +6,7 @@
 > 代码：[https://github.com/WangCan1178/ExpG](https://github.com/WangCan1178/ExpG)

 <p align="center">
-  <img src="gitcha.png" alt="ExpG 挑战与概览" width="85%">
+  <img src="./gitcha.png" alt="ExpG 挑战与概览" width="85%">
 </p>

 ### 简介
--- a/docs/.vitepress/config.mts
+++ b/docs/.vitepress/config.mts
@ -0,0 +1,342 @@
+import fs from "node:fs";
+import path from "node:path";
+import { execFileSync } from "node:child_process";
+import { fileURLToPath } from "node:url";
+import { defineConfig, type DefaultTheme } from "vitepress";
+import { legacyRoutes } from "./legacy-routes.mjs";
+
+const sourceRoot = path.resolve(path.dirname(fileURLToPath(import.meta.url)), "..");
+const repositoryRoot = path.resolve(sourceRoot, "../../..");
+const repository = "https://github.com/agentscope-ai/ReMe";
+const base = process.env.DOCS_BASE || "/";
+
+function readSourceMap(): Record<string, string> {
+  try {
+    return JSON.parse(fs.readFileSync(path.join(sourceRoot, ".source-map.json"), "utf8"));
+  } catch {
+    return {};
+  }
+}
+
+const sourceMap = readSourceMap();
+
+function collectMarkdown(directory: string, root = directory): string[] {
+  const files: string[] = [];
+  for (const entry of fs.readdirSync(directory, { withFileTypes: true })) {
+    if (entry.name.startsWith(".") || entry.name === "public" || entry.name === "figure") continue;
+    const absolute = path.join(directory, entry.name);
+    if (entry.isDirectory()) files.push(...collectMarkdown(absolute, root));
+    else if (entry.name.endsWith(".md")) files.push(path.relative(root, absolute).replaceAll(path.sep, "/"));
+  }
+  return files.sort();
+}
+
+function buildLlmsFiles(outDir: string) {
+  const pages = collectMarkdown(sourceRoot);
+  const index = [
+    "# ReMe Documentation",
+    "",
+    "> Local-first, file-native memory for agents.",
+    "",
+    ...pages.map((relativePath) => {
+      const source = fs.readFileSync(path.join(sourceRoot, relativePath), "utf8");
+      const title = source.match(/^#\s+(.+)$/m)?.[1]
+        || source.match(/^title:\s*(.+)$/m)?.[1]
+        || path.basename(relativePath, ".md");
+      const route = relativePath.replace(/(?:^|\/)index\.md$/, "").replace(/\.md$/, "");
+      return `- [${title}](https://reme.agentscope.io/${route})`;
+    }),
+    "",
+  ];
+  fs.writeFileSync(path.join(outDir, "llms.txt"), index.join("\n"), "utf8");
+
+  const full = ["# ReMe Documentation", ""];
+  for (const relativePath of pages) {
+    const source = fs.readFileSync(path.join(sourceRoot, relativePath), "utf8");
+    full.push(`<!-- source: ${sourcePathFor(relativePath)} -->`, "", source, "", "---", "");
+    const pageDir = path.join(outDir, relativePath.replace(/\.md$/, ""));
+    fs.mkdirSync(pageDir, { recursive: true });
+    fs.writeFileSync(path.join(pageDir, "llms.txt"), source, "utf8");
+  }
+  fs.writeFileSync(path.join(outDir, "llms-full.txt"), full.join("\n"), "utf8");
+}
+
+function sourcePathFor(relativePath: string) {
+  return sourceMap[relativePath] || `docs/${relativePath}`;
+}
+
+function sourceLastUpdated(relativePath: string): number | undefined {
+  const sourcePath = sourcePathFor(relativePath);
+  try {
+    const timestamp = execFileSync("git", ["log", "-1", "--format=%ct", "--", sourcePath], {
+      cwd: repositoryRoot,
+      encoding: "utf8",
+    }).trim();
+    if (timestamp) return Number(timestamp) * 1000;
+  } catch {
+    // Fall back to the canonical file timestamp outside a Git checkout.
+  }
+  try {
+    return fs.statSync(path.join(repositoryRoot, sourcePath)).mtimeMs;
+  } catch {
+    return undefined;
+  }
+}
+
+const legacyRedirectScript = `(() => {
+  const routes = ${JSON.stringify(legacyRoutes)};
+  const id = new URLSearchParams(window.location.search).get("doc");
+  const target = id && routes[id];
+  const base = ${JSON.stringify(base)};
+  if (target) {
+    const destination = /^https?:/.test(target)
+      ? target
+      : base.replace(/\\/$/, "") + target;
+    window.location.replace(destination + window.location.hash);
+    return;
+  }
+  const root = base.endsWith("/") ? base : base + "/";
+  if (window.location.pathname === root) {
+    window.location.replace(root + "zh/" + window.location.hash);
+  }
+})();`;
+
+function nav(language: "zh" | "en"): DefaultTheme.NavItem[] {
+  const zh = language === "zh";
+  return [
+    { text: zh ? "开始使用" : "Get Started", link: `/${language}/quick_start` },
+    { text: zh ? "核心概念" : "Concepts", link: `/${language}/memory_as_file` },
+    { text: zh ? "指南" : "Guides", link: `/${language}/auto_memory` },
+    { text: zh ? "集成" : "Integrations", link: `/${language}/integrations` },
+    { text: zh ? "API 参考" : "API Reference", link: `/${language}/reference/cli` },
+    { text: zh ? "常见问题" : "FAQ", link: `/${language}/faq` },
+  ];
+}
+
+function sidebar(language: "zh" | "en"): DefaultTheme.SidebarItem[] {
+  const zh = language === "zh";
+  return [
+    {
+      text: zh ? "开始使用" : "Get Started",
+      collapsed: false,
+      items: [
+        { text: zh ? "项目介绍" : "Introduction", link: `/${language}/` },
+        { text: zh ? "快速开始" : "Quick Start", link: `/${language}/quick_start` },
+        { text: zh ? "基础配置" : "Configuration", link: `/${language}/configuration` },
+        { text: zh ? "服务与部署" : "Services and Deployment", link: `/${language}/services` },
+      ],
+    },
+    {
+      text: zh ? "核心概念" : "Core Concepts",
+      collapsed: false,
+      items: [
+        { text: zh ? "文件即记忆" : "Memory as File", link: `/${language}/memory_as_file` },
+        { text: zh ? "记忆检索" : "Memory Search", link: `/${language}/memory_search` },
+        { text: zh ? "自动关联" : "Auto Link", link: `/${language}/auto_link` },
+        { text: zh ? "应用场景" : "Application Scenarios", link: `/${language}/reme_scene` },
+      ],
+    },
+    {
+      text: zh ? "记忆工作流" : "Memory Workflows",
+      collapsed: false,
+      items: [
+        { text: "Auto Memory", link: `/${language}/auto_memory` },
+        { text: "Auto Resource", link: `/${language}/auto_resource` },
+        { text: "Auto Dream", link: `/${language}/auto_dream` },
+        { text: "Proactive", link: `/${language}/proactive` },
+      ],
+    },
+    {
+      text: zh ? "Agent 集成" : "Agent Integrations",
+      collapsed: true,
+      items: [
+        { text: zh ? "集成总览" : "Overview", link: `/${language}/integrations` },
+        { text: "Claude Code", link: `/${language}/integrations/claude-code` },
+        { text: "Hermes Agent", link: `/${language}/integrations/hermes` },
+        { text: "TypeScript / OpenClaw / DSH", link: `/${language}/integrations/typescript` },
+      ],
+    },
+    {
+      text: zh ? "工作区与插件" : "Workspace and Plugins",
+      collapsed: true,
+      items: [
+        { text: "ReMe Studio", link: `/${language}/workspace/studio` },
+        { text: zh ? "插件管理" : "Plugin Management", link: `/${language}/plugin_management` },
+        { text: zh ? "插件开发" : "Plugin Development", link: `/${language}/plugin_development` },
+        { text: zh ? "每日论文" : "Daily Paper", link: `/${language}/plugins/daily-paper` },
+        { text: "Auto Fin", link: `/${language}/plugins/auto-fin` },
+        { text: "LME", link: `/${language}/plugins/lme` },
+        { text: "BEAM", link: `/${language}/plugins/beam` },
+      ],
+    },
+    {
+      text: zh ? "API 参考" : "API Reference",
+      collapsed: true,
+      items: [
+        { text: "CLI", link: `/${language}/reference/cli` },
+        { text: zh ? "Job API" : "Job API", link: `/${language}/reference/jobs` },
+        { text: "HTTP / MCP", link: `/${language}/services#http-api` },
+      ],
+    },
+    {
+      text: zh ? "运维" : "Operations",
+      collapsed: true,
+      items: [
+        { text: zh ? "诊断、备份与恢复" : "Diagnostics, Backup, and Recovery", link: `/${language}/operations` },
+        { text: zh ? "常见问题" : "FAQ", link: `/${language}/faq` },
+      ],
+    },
+    {
+      text: zh ? "开发者" : "Development",
+      collapsed: true,
+      items: [
+        { text: zh ? "代码框架" : "Framework", link: `/${language}/framework` },
+        { text: zh ? "开源与贡献" : "Contributing", link: `/${language}/contributing` },
+      ],
+    },
+    {
+      text: zh ? "评测" : "Benchmarks",
+      collapsed: true,
+      items: [
+        { text: "BEAM", link: `/${language}/benchmarks/beam` },
+        { text: "LongMemEval", link: `/${language}/benchmarks/longmemeval` },
+        { text: "π-Bench", link: `/${language}/benchmarks/pibench` },
+        { text: "Tool Memory / ExpG", link: `/${language}/benchmarks/toolmemory` },
+      ],
+    },
+  ];
+}
+
+function configureRepositoryLinks(md: any) {
+  for (const ruleName of ["link_open", "image"] as const) {
+    const original = md.renderer.rules[ruleName];
+    md.renderer.rules[ruleName] = (tokens: any[], index: number, options: any, env: any, self: any) => {
+      const attribute = ruleName === "image" ? "src" : "href";
+      const token = tokens[index];
+      const attributeIndex = token.attrIndex(attribute);
+      const target = attributeIndex >= 0 ? token.attrs[attributeIndex][1] : "";
+      if (target && !/^(?:[a-z]+:|#|\/)/i.test(target)) {
+        const cleanTarget = target.split("#")[0].split("?")[0];
+        const generatedTarget = path.resolve(sourceRoot, path.dirname(env.relativePath), cleanTarget);
+        if (!fs.existsSync(generatedTarget)) {
+          const originalPage = sourcePathFor(env.relativePath);
+          const originalTarget = path.posix.normalize(path.posix.join(path.posix.dirname(originalPage), cleanTarget));
+          const suffix = target.slice(cleanTarget.length);
+          const url = ruleName === "image"
+            ? `https://raw.githubusercontent.com/agentscope-ai/ReMe/main/${originalTarget}${suffix}`
+            : `${repository}/blob/main/${originalTarget}${suffix}`;
+          token.attrs[attributeIndex][1] = url;
+        }
+      }
+      return original ? original(tokens, index, options, env, self) : self.renderToken(tokens, index, options);
+    };
+  }
+}
+
+export default defineConfig({
+  lang: "zh-CN",
+  title: "ReMe",
+  description: "Local-first, file-native memory for agents",
+  base,
+  cleanUrls: true,
+  lastUpdated: true,
+  ignoreDeadLinks: [/^http:\/\/localhost(?::\d+)?(?:\/|$)/],
+  sitemap: {
+    hostname: "https://reme.agentscope.io",
+    transformItems(items) {
+      const isRoot = (url: string) => url.replace(/^\/+|\/+$/g, "") === "";
+      return items.filter((item) => !isRoot(item.url)).map((item) => {
+        const route = item.url.replace(/^\/+/, "");
+        const relativePath = !route || route.endsWith("/") ? `${route}index.md` : `${route}.md`;
+        const links = item.links?.filter((link) => !isRoot(link.url));
+        return { ...item, links, lastmod: sourceLastUpdated(relativePath) };
+      });
+    },
+  },
+  head: [
+    ["link", { rel: "icon", type: "image/svg+xml", href: `${base}favicon.svg` }],
+    ["meta", { name: "theme-color", content: "#087f6a" }],
+    ["script", {}, legacyRedirectScript],
+  ],
+  markdown: {
+    config: configureRepositoryLinks,
+  },
+  transformPageData(pageData, { siteConfig }) {
+    const sourcePath = path.join(siteConfig.srcDir, pageData.relativePath);
+    pageData.frontmatter._sourcePath = sourcePathFor(pageData.relativePath);
+    pageData.lastUpdated = sourceLastUpdated(pageData.relativePath);
+    try {
+      pageData.frontmatter._rawMarkdown = fs.readFileSync(sourcePath, "utf8");
+    } catch {
+      pageData.frontmatter._rawMarkdown = "";
+    }
+  },
+  buildEnd(siteConfig) {
+    buildLlmsFiles(siteConfig.outDir);
+  },
+  themeConfig: {
+    logo: "/favicon.svg",
+    siteTitle: "ReMe",
+    nav: [
+      ...nav("zh"),
+      {
+        text: "语言",
+        items: [
+          { text: "简体中文", link: "/zh/" },
+          { text: "English", link: "/en/" },
+        ],
+      },
+    ],
+    outline: { label: "页面导航", level: [2, 3] },
+    search: {
+      provider: "local",
+      options: {
+        locales: {
+          zh: {
+            translations: {
+              button: { buttonText: "搜索文档", buttonAriaLabel: "搜索文档" },
+              modal: {
+                noResultsText: "没有找到相关内容",
+                resetButtonTitle: "清除查询",
+                footer: { selectText: "选择", navigateText: "切换", closeText: "关闭" },
+              },
+            },
+          },
+        },
+      },
+    },
+    socialLinks: [{ icon: "github", link: repository }],
+    footer: {
+      message: "Released under the Apache-2.0 License.",
+      copyright: "Copyright ReMe contributors",
+    },
+  },
+  locales: {
+    zh: {
+      label: "简体中文",
+      lang: "zh-CN",
+      link: "/zh/",
+      themeConfig: {
+        nav: nav("zh"),
+        sidebar: { "/zh/": sidebar("zh") },
+        outline: { label: "页面导航", level: [2, 3] },
+        docFooter: { prev: "上一页", next: "下一页" },
+        darkModeSwitchLabel: "外观",
+        sidebarMenuLabel: "菜单",
+        returnToTopLabel: "返回顶部",
+        langMenuLabel: "切换语言",
+      },
+    },
+    en: {
+      label: "English",
+      lang: "en-US",
+      link: "/en/",
+      themeConfig: {
+        nav: nav("en"),
+        sidebar: { "/en/": sidebar("en") },
+        outline: { label: "On this page", level: [2, 3] },
+        docFooter: { prev: "Previous page", next: "Next page" },
+      },
+    },
+  },
+});
--- a/docs/.vitepress/legacy-routes.mjs
+++ b/docs/.vitepress/legacy-routes.mjs
@ -0,0 +1,47 @@
+export const legacyRoutes = {
+  "readme-zh": "/zh/",
+  "readme-en": "/en/",
+  "zh-quick_start": "/zh/quick_start",
+  "en-quick_start": "/en/quick_start",
+  "zh-plugin_management": "/zh/plugin_management",
+  "en-plugin_management": "/en/plugin_management",
+  "zh-memory_as_file": "/zh/memory_as_file",
+  "en-memory_as_file": "/en/memory_as_file",
+  "zh-memory_search": "/zh/memory_search",
+  "en-memory_search": "/en/memory_search",
+  "zh-auto_memory": "/zh/auto_memory",
+  "en-auto_memory": "/en/auto_memory",
+  "zh-auto_resource": "/zh/auto_resource",
+  "en-auto_resource": "/en/auto_resource",
+  "zh-auto_link": "/zh/auto_link",
+  "en-auto_link": "/en/auto_link",
+  "zh-auto_dream": "/zh/auto_dream",
+  "en-auto_dream": "/en/auto_dream",
+  "zh-proactive": "/zh/proactive",
+  "en-proactive": "/en/proactive",
+  "zh-reme_scene": "/zh/reme_scene",
+  "en-reme_scene": "/en/reme_scene",
+  "zh-framework": "/zh/framework",
+  "en-framework": "/en/framework",
+  "zh-reme-blog": "/zh/reme-blog",
+  "en-reme-blog": "/en/reme-blog",
+  "zh-contributing": "/zh/contributing",
+  "en-contributing": "/en/contributing",
+  "typescript-zh": "/zh/integrations/typescript",
+  "typescript-en": "/en/integrations/typescript",
+  "studio-zh": "/zh/workspace/studio",
+  "studio-en": "/en/workspace/studio",
+  "daily-paper-zh": "/zh/plugins/daily-paper",
+  "daily-paper-en": "/en/plugins/daily-paper",
+  "auto-fin-zh": "/zh/plugins/auto-fin",
+  "auto-fin-en": "/en/plugins/auto-fin",
+  "beam-zh": "/zh/benchmarks/beam",
+  "beam-en": "/en/benchmarks/beam",
+  "longmemeval-zh": "/zh/benchmarks/longmemeval",
+  "longmemeval-en": "/en/benchmarks/longmemeval",
+  "pibench-zh": "/zh/benchmarks/pibench",
+  "pibench-en": "/en/benchmarks/pibench",
+  "toolmemory-zh": "/zh/benchmarks/toolmemory",
+  "toolmemory-en": "/en/benchmarks/toolmemory",
+  "agents-guide": "https://github.com/agentscope-ai/ReMe/blob/main/AGENTS.md",
+};
--- a/docs/.vitepress/theme/CopyMarkdownButton.vue
+++ b/docs/.vitepress/theme/CopyMarkdownButton.vue
@ -0,0 +1,32 @@
+<script setup lang="ts">
+import { computed, ref } from "vue";
+import { useData } from "vitepress";
+
+const { frontmatter, lang } = useData();
+const copied = ref(false);
+const label = computed(() => {
+  if (copied.value) return lang.value.startsWith("zh") ? "已复制" : "Copied";
+  return lang.value.startsWith("zh") ? "复制 Markdown" : "Copy Markdown";
+});
+
+async function copyMarkdown() {
+  const markdown = String(frontmatter.value._rawMarkdown || "");
+  if (!markdown) return;
+  await navigator.clipboard.writeText(markdown);
+  copied.value = true;
+  window.setTimeout(() => { copied.value = false; }, 1800);
+}
+</script>
+
+<template>
+  <div class="copy-markdown-wrap">
+    <button class="copy-markdown" type="button" :class="{ copied }" @click="copyMarkdown">
+      <svg v-if="!copied" viewBox="0 0 24 24" aria-hidden="true">
+        <rect x="9" y="9" width="13" height="13" rx="2" />
+        <path d="M5 15H4a2 2 0 0 1-2-2V4a2 2 0 0 1 2-2h9a2 2 0 0 1 2 2v1" />
+      </svg>
+      <svg v-else viewBox="0 0 24 24" aria-hidden="true"><path d="m5 12 4 4L19 6" /></svg>
+      {{ label }}
+    </button>
+  </div>
+</template>
--- a/docs/.vitepress/theme/SourceLink.vue
+++ b/docs/.vitepress/theme/SourceLink.vue
@ -0,0 +1,15 @@
+<script setup lang="ts">
+import { computed } from "vue";
+import { useData } from "vitepress";
+
+const { frontmatter, lang } = useData();
+const sourcePath = computed(() => String(frontmatter.value._sourcePath || ""));
+const label = computed(() => lang.value.startsWith("zh") ? "在 GitHub 查看源文件" : "View source on GitHub");
+const href = computed(() => `https://github.com/agentscope-ai/ReMe/blob/main/${sourcePath.value}`);
+</script>
+
+<template>
+  <div v-if="sourcePath" class="source-link-wrap">
+    <a :href="href" target="_blank" rel="noreferrer">{{ label }} ↗</a>
+  </div>
+</template>
--- a/docs/.vitepress/theme/custom.css
+++ b/docs/.vitepress/theme/custom.css
@ -0,0 +1,224 @@
+:root {
+  --vp-layout-max-width: 1560px;
+  --vp-c-brand-1: #087f6a;
+  --vp-c-brand-2: #086554;
+  --vp-c-brand-3: #19a98f;
+  --vp-c-brand-soft: rgba(8, 127, 106, 0.14);
+  --vp-c-bg: #ffffff;
+  --vp-c-bg-alt: #f4f7f5;
+  --vp-c-bg-elv: #ffffff;
+  --vp-c-bg-soft: #f1f6f3;
+  --vp-c-text-1: #17221d;
+  --vp-c-text-2: #526159;
+  --vp-c-text-3: #718078;
+  --vp-c-divider: #dce5e0;
+  --vp-font-family-base: Inter, ui-sans-serif, system-ui, -apple-system, BlinkMacSystemFont, "Segoe UI", sans-serif;
+  --vp-font-family-mono: "SFMono-Regular", Consolas, "Liberation Mono", monospace;
+  --reme-blue: #3156d9;
+  --reme-green: #087f6a;
+}
+
+.dark {
+  --vp-c-brand-1: #57dfc3;
+  --vp-c-brand-2: #35c6a9;
+  --vp-c-brand-3: #087f6a;
+  --vp-c-brand-soft: rgba(87, 223, 195, 0.14);
+  --vp-c-bg: #0d1512;
+  --vp-c-bg-alt: #09100d;
+  --vp-c-bg-elv: #14201b;
+  --vp-c-bg-soft: #17251f;
+  --vp-c-text-1: #edf7f3;
+  --vp-c-text-2: #bacbc4;
+  --vp-c-text-3: #91a49c;
+  --vp-c-divider: #283a33;
+}
+
+body {
+  background:
+    radial-gradient(circle at 8% 8%, rgba(25, 201, 176, 0.055), transparent 28rem),
+    var(--vp-c-bg);
+}
+
+.VPNav {
+  border-bottom: 1px solid color-mix(in srgb, var(--vp-c-divider) 84%, transparent);
+  background: color-mix(in srgb, var(--vp-c-bg) 88%, transparent);
+  backdrop-filter: blur(18px) saturate(140%);
+}
+
+.VPNavBarTitle .logo {
+  width: 30px;
+  height: 30px;
+}
+
+.VPNavBarTitle .title {
+  font-weight: 780;
+  letter-spacing: -0.02em;
+}
+
+.VPNavBarSearch .DocSearch-Button,
+.VPNavBarSearch button {
+  min-width: 190px;
+  border: 1px solid var(--vp-c-divider);
+  border-radius: 10px;
+  background: var(--vp-c-bg-alt);
+}
+
+.VPSidebar {
+  border-right: 1px solid var(--vp-c-divider);
+  background: color-mix(in srgb, var(--vp-c-bg-alt) 82%, transparent);
+}
+
+.VPSidebarItem .text {
+  font-size: 14px;
+}
+
+.VPSidebarItem.level-0 > .item > .text {
+  color: var(--vp-c-text-1);
+  font-weight: 750;
+}
+
+.VPSidebarItem.is-active > .item .link > .text {
+  color: var(--vp-c-brand-1);
+}
+
+.VPDocAsideOutline {
+  border-left-color: var(--vp-c-divider);
+}
+
+.VPDoc .container > .content {
+  min-width: 0;
+}
+
+.VPDoc .content-container {
+  max-width: 900px !important;
+}
+
+.vp-doc {
+  color: var(--vp-c-text-1);
+  font-size: 16px;
+  line-height: 1.78;
+}
+
+.vp-doc h1 {
+  margin-bottom: 28px;
+  font-size: clamp(36px, 5vw, 52px);
+  line-height: 1.08;
+  letter-spacing: -0.045em;
+}
+
+.vp-doc h2 {
+  margin-top: 52px;
+  border-top-color: var(--vp-c-divider);
+  font-size: 27px;
+  letter-spacing: -0.025em;
+}
+
+.vp-doc h3 {
+  margin-top: 34px;
+  font-size: 20px;
+}
+
+.vp-doc :not(pre) > code {
+  border-radius: 5px;
+  color: color-mix(in srgb, var(--vp-c-brand-1) 82%, var(--vp-c-text-1));
+}
+
+.vp-doc div[class*="language-"] {
+  border: 1px solid var(--vp-c-divider);
+  border-radius: 12px;
+  box-shadow: inset 3px 0 0 color-mix(in srgb, var(--vp-c-brand-1) 62%, transparent);
+}
+
+.copy-markdown-wrap {
+  display: flex;
+  justify-content: flex-end;
+  margin-bottom: 18px;
+}
+
+.copy-markdown {
+  display: inline-flex;
+  align-items: center;
+  gap: 7px;
+  min-height: 34px;
+  padding: 6px 12px;
+  border: 1px solid var(--vp-c-divider);
+  border-radius: 9px;
+  color: var(--vp-c-text-2);
+  background: var(--vp-c-bg-soft);
+  cursor: pointer;
+  font-size: 13px;
+  font-weight: 650;
+}
+
+.copy-markdown:hover,
+.copy-markdown.copied {
+  border-color: var(--vp-c-brand-1);
+  color: var(--vp-c-brand-1);
+}
+
+.copy-markdown svg {
+  width: 15px;
+  height: 15px;
+  fill: none;
+  stroke: currentColor;
+  stroke-linecap: round;
+  stroke-linejoin: round;
+  stroke-width: 2;
+}
+
+.source-link-wrap {
+  margin-top: 48px;
+  padding-top: 20px;
+  border-top: 1px solid var(--vp-c-divider);
+  font-size: 13px;
+}
+
+.source-link-wrap a {
+  color: var(--vp-c-text-3);
+  text-decoration: none;
+}
+
+.source-link-wrap a:hover {
+  color: var(--vp-c-brand-1);
+}
+
+.VPHome {
+  background:
+    radial-gradient(circle at 14% 14%, rgba(25, 201, 176, 0.14), transparent 30rem),
+    radial-gradient(circle at 86% 10%, rgba(49, 86, 217, 0.11), transparent 28rem);
+}
+
+.VPHero .name {
+  background: linear-gradient(120deg, var(--reme-green), var(--reme-blue));
+  background-clip: text;
+  -webkit-background-clip: text;
+  -webkit-text-fill-color: transparent;
+}
+
+.VPHero .name,
+.VPHero .text {
+  letter-spacing: -0.05em;
+}
+
+.VPHero .image-bg {
+  background: linear-gradient(135deg, rgba(25, 201, 176, 0.28), rgba(49, 86, 217, 0.2));
+  filter: blur(46px);
+}
+
+.VPFeature {
+  border-color: color-mix(in srgb, var(--vp-c-brand-1) 18%, var(--vp-c-divider));
+  border-radius: 16px;
+  background: linear-gradient(150deg, var(--vp-c-bg-elv), var(--vp-c-bg-soft));
+  transition: transform 160ms ease, box-shadow 160ms ease;
+}
+
+.VPFeature:hover {
+  transform: translateY(-3px);
+  box-shadow: 0 16px 34px rgba(21, 58, 43, 0.1);
+}
+
+@media (max-width: 768px) {
+  .vp-doc h1 { font-size: 34px; }
+  .vp-doc h2 { margin-top: 44px; font-size: 24px; }
+  .copy-markdown-wrap { margin-top: -8px; }
+}
--- a/docs/.vitepress/theme/index.ts
+++ b/docs/.vitepress/theme/index.ts
@ -0,0 +1,15 @@
+import { h } from "vue";
+import DefaultTheme from "vitepress/theme";
+import CopyMarkdownButton from "./CopyMarkdownButton.vue";
+import SourceLink from "./SourceLink.vue";
+import "./custom.css";
+
+export default {
+  extends: DefaultTheme,
+  Layout() {
+    return h(DefaultTheme.Layout, null, {
+      "doc-before": () => h(CopyMarkdownButton),
+      "doc-after": () => h(SourceLink),
+    });
+  },
+};
--- a/docs/en/configuration.md
+++ b/docs/en/configuration.md
@ -0,0 +1,157 @@
+---
+title: Configuration
+description: ReMe configuration files, environment expansion, command-line overrides, and core components.
+---
+
+# Configuration
+
+ReMe uses YAML or JSON to describe its Service, Jobs, and Components. The built-in default is `reme/config/default.yaml`. Select another configuration at startup and apply command-line overrides when needed.
+
+## Precedence
+
+Configuration is merged in this order, with later values winning:
+
+1. `application_defaults` from enabled plugins.
+2. The selected file; `default` is used when none is specified.
+3. CLI dot-notation overrides.
+
+```bash
+reme start
+reme start config=demo
+reme start config=/absolute/path/to/app.yaml
+reme start service.port=8181 workspace_dir=/data/reme
+```
+
+`config` accepts a built-in name or a `.yaml`, `.yml`, or `.json` file. Overrides are deep-merged, so changing `service.port` preserves sibling service settings.
+
+## CLI values
+
+Arguments use `key=value`; leading `-` or `--` is accepted:
+
+```bash
+reme start --service.port=8181 --service.web_enabled=false
+```
+
+Values support null, booleans, numbers, JSON arrays and objects, quoted JSON strings, and plain strings. Numeric-looking values with leading zeroes, such as `007`, remain strings. Quote values such as `"true"` in JSON when they must remain strings.
+
+## Environment variables
+
+Configuration recursively expands:
+
+```yaml
+api_key: ${LLM_API_KEY}
+base_url: ${LLM_BASE_URL:-https://example.com/v1}
+```
+
+`${VAR}` fails when undefined; `${VAR:-default}` uses its fallback. ReMe also searches for `.env` from the command's working directory through at most five parents.
+
+Keep secrets in `.env` or the process environment, never in committed configuration.
+
+## Application fields
+
+| Field | Default | Purpose |
+|---|---|---|
+| `app_name` | `ReMe` | Display name |
+| `workspace_dir` | `.reme` | User-owned workspace root, normalized to an absolute path |
+| `metadata_dir` | `metadata` | Rebuildable indexes, graphs, and catalogs |
+| `session_dir` | `session` | Agent sessions; standard transcripts use `session/dialog` |
+| `mem_session_dir` | `mem_session` | Agent-wrapper sessions and configuration |
+| `resource_dir` | `resource` | External resources |
+| `daily_dir` | `daily` | Daily memory |
+| `digest_dir` | `digest` | Consolidated long-term memory |
+| `timezone` | `Asia/Shanghai` | IANA timezone used for dates and cron jobs |
+| `language` | empty | Default language for LLM interactions |
+| `plugins` | `[]` | Installed plugins enabled for this Application |
+| `service` | HTTP | Service configuration |
+| `jobs` | default Jobs | Job configurations by name |
+| `components` | defaults | Components grouped by type and name |
+
+`session_dir` must remain workspace-relative.
+
+## LLM
+
+The default LLM uses an OpenAI-compatible interface:
+
+```yaml
+components:
+  as_llm:
+    default:
+      backend: openai
+      model: qwen3.7-plus
+      context_size: 200000
+      credential:
+        api_key: ${LLM_API_KEY:-}
+        base_url: ${LLM_BASE_URL:-}
+```
+
+Built-in registrations include `openai`, `anthropic`, `dashscope`, `deepseek`, `gemini`, `moonshot`, `ollama`, and `xai`. Their detailed model fields follow the corresponding AgentScope wrappers.
+
+File operations, BM25 search, and wikilink traversal do not require an LLM. Evolution workflows such as `auto_memory`, `auto_resource`, and `auto_dream` do.
+
+## Embeddings
+
+Vector retrieval is disabled by default. Credentials alone do not enable it: configure `as_embedding`, `embedding_store`, and connect the store to `file_store`.
+
+```yaml
+components:
+  as_embedding:
+    default:
+      backend: openai
+      model: text-embedding-v4
+      dimensions: 1024
+      credential:
+        api_key: ${EMBEDDING_API_KEY}
+        base_url: ${EMBEDDING_BASE_URL:-https://dashscope.aliyuncs.com/compatible-mode/v1}
+  embedding_store:
+    default:
+      backend: local
+      as_embedding: default
+  file_store:
+    default:
+      backend: local
+      embedding_store: default
+      keyword_index: default
+      file_graph: default
+```
+
+Rebuild the embedding index after changing the model or dimensions.
+
+## Service and Jobs
+
+Minimal HTTP configuration:
+
+```yaml
+service:
+  backend: http
+  host: 127.0.0.1
+  port: 2333
+  web_enabled: true
+  mcp_enabled: true
+  mcp_path: /mcp
+```
+
+A Job declares a backend, parameter schema, and ordered Steps:
+
+```yaml
+jobs:
+  example:
+    backend: base
+    description: Example job
+    parameters:
+      type: object
+      properties:
+        text: { type: string }
+      required: [text]
+    steps:
+      - backend: example_step
+```
+
+Set `enable_serve: false` to keep a Job internal. Background and cron Jobs are never service-exposed.
+
+## Inspect the effective configuration
+
+```bash
+reme app_config
+```
+
+The result is the merged, validated configuration with secrets redacted. Use it when diagnosing plugin or override precedence. The authoritative contracts remain `reme/schema/application_config.py` and `reme/config/default.yaml`.
--- a/docs/en/contributing.md
+++ b/docs/en/contributing.md
@ -209,6 +209,13 @@ Documentation lives under:
 docs/
 ```

+User guides should have matching `docs/zh/` and `docs/en/` versions and appear in the corresponding navigation in
+`docs/.vitepress/config.mts`. The ReMe Studio, TypeScript, plugin, and benchmark READMEs remain canonical in their own
+directories; `github-pages/scripts/generate-content.mjs` mirrors them during builds. Never edit `.generated/` or `dist/`.
+
+The Job API reference is generated from `reme/config/default.yaml`. Update that YAML and its tests when a default Job
+contract changes rather than editing generated pages.
+
 Documentation should:

 - Use clear titles that directly identify a capability or flow.
@ -216,6 +223,15 @@ Documentation should:
 - Use real repository paths such as `reme/config/default.yaml`, `reme/steps/`, and `tests/unit/`.
 - Describe default behavior according to the current code, `pyproject.toml`, and default configuration.

+Validate the documentation site with:
+
+```bash
+cd github-pages
+npm ci
+npm test
+npm run build
+```
+
 ---

 ## Getting Help
--- a/docs/en/faq.md
+++ b/docs/en/faq.md
@ -0,0 +1,71 @@
+---
+title: Frequently Asked Questions
+description: Quick answers for ReMe installation, services, models, retrieval, files, and plugins.
+---
+
+# Frequently Asked Questions
+
+## Do basic file operations require a model API key?
+
+No. `write`, `read`, `list`, `stat`, BM25 search, and wikilink traversal work without model credentials. `auto_memory`, `auto_resource`, and `auto_dream` require an LLM.
+
+## Why is search still BM25-only after setting an embedding key?
+
+Embeddings are disabled by default. Configure `as_embedding` and `embedding_store`, then connect `file_store.default.embedding_store` to that component. See [Configuration](./configuration.md#embeddings).
+
+## Why did `reme reindex` not discover a new file?
+
+`reindex` rebuilds indexes from current `file_chunks`; it does not scan the workspace. Check `index_update_loop`, the watched directory and extension, and `health_check`.
+
+## How do I use another workspace?
+
+```bash
+reme start workspace_dir=/absolute/path/to/memory
+```
+
+Ordinary CLI calls discover the running service, so they do not need the workspace argument again.
+
+## What if port 2333 is occupied?
+
+Do not stop an unknown listener. Select another port:
+
+```bash
+reme start service.port=8181
+```
+
+Then confirm it with `reme find_reme`.
+
+## Why is an installed plugin missing its Jobs?
+
+Installation only makes the distribution discoverable in the active Python environment. Enable it for the Application:
+
+```bash
+reme start plugins='["auto-fin"]'
+```
+
+Restart a running service after changing package or enablement state.
+
+## May I edit workspace Markdown directly?
+
+Yes. Files are the source of truth and watchers ingest changes. Keep frontmatter valid, use complete workspace-relative wikilinks, and avoid unconditional concurrent saves.
+
+## May I expose ReMe publicly?
+
+Not with the default configuration alone. Jobs can write and delete, HTTP CORS is permissive, and there is no general authentication layer. Use a controlled network or authenticated TLS reverse proxy and restrict `service.jobs`.
+
+## How should I back up and migrate memory?
+
+Stop writes and back up the complete workspace. `session/`, `resource/`, `daily/`, and `digest/` are the key sources; `metadata/` can be backed up or rebuilt. See [Diagnostics, Backup, and Recovery](./operations.md).
+
+## Why is Studio unavailable?
+
+The base `reme-ai` package has no frontend assets. Install `reme-ai[web]` or `reme-ai[core]`, or set `service.web_static_dir`. Missing Studio assets do not disable the Job API.
+
+## Which capabilities does the running service expose?
+
+```bash
+reme help
+reme app_config
+```
+
+Static documentation describes defaults; plugins and custom configuration may change the active service.
--- a/docs/en/index.md
+++ b/docs/en/index.md
@ -0,0 +1,36 @@
+---
+layout: home
+title: ReMe Documentation
+titleTemplate: false
+hero:
+  name: ReMe
+  text: Memory that works for agents
+  tagline: Files remain yours. ReMe turns conversations and resources into readable, editable, searchable, interconnected local files.
+  image:
+    src: /favicon.svg
+    alt: ReMe
+  actions:
+    - theme: brand
+      text: Quick Start
+      link: /en/quick_start
+    - theme: alt
+      text: Configuration
+      link: /en/configuration
+features:
+  - icon: 📁
+    title: Memory as files
+    details: Workspace files are the durable source of truth; indexes, graphs, and caches remain rebuildable.
+    link: /en/memory_as_file
+  - icon: 🧠
+    title: Memory workflows
+    details: Turn conversations and resources into daily notes, then consolidate them into connected long-term digests.
+    link: /en/auto_memory
+  - icon: 🔎
+    title: Search and graph
+    details: Combine keyword and optional vector retrieval with wikilink graph expansion.
+    link: /en/memory_search
+  - icon: 🔌
+    title: Agent integrations
+    details: Connect existing agents through the CLI, HTTP, MCP, and host adapters.
+    link: /en/integrations
+---
--- a/docs/en/integrations.md
+++ b/docs/en/integrations.md
@ -0,0 +1,63 @@
+---
+title: Agent Integrations
+description: Connect ReMe to agents through the CLI, HTTP, MCP, Skills, and host adapters.
+---
+
+# Agent Integrations
+
+ReMe keeps memory in an independent service and a user-owned workspace. Multiple agents can call the same memory system without binding storage to one model or host.
+
+## Choose an interface
+
+| Scenario | Recommended interface |
+|---|---|
+| Local script or hook | ReMe CLI |
+| Application backend | HTTP Client |
+| Tool-protocol host | MCP |
+| TypeScript agent | `@agentscope-ai/reme` |
+| Claude Code | MCP + Skill + Stop Hook |
+| Hermes Agent | Memory provider adapter |
+| Codex or another coding agent | `reme_memory` Skill or MCP |
+
+## General memory loop
+
+1. Before answering, call `search` for relevant memory.
+2. Use `read` on high-value results and `traverse` when relationships matter.
+3. Retain workspace-relative source paths in the answer.
+4. At session end, pass source messages to `auto_memory`.
+5. Let background or scheduled workflows consolidate daily notes into digest memory.
+
+An empty search result must remain empty; do not present model inference as recalled history.
+
+## MCP
+
+The default HTTP service exposes streamable HTTP MCP at `http://127.0.0.1:2333/mcp`. Common tools include `search`, `read`, `traverse`, `list`, `auto_memory`, and `proactive`.
+
+Use `service.jobs` to expose a read-only subset or keep write tools in a separate configuration.
+
+## CLI and Skill
+
+`skills/reme_memory/SKILL.md` defines a general workflow for agents that can run local commands: installation checks, service discovery, retrieval, reading, and persistence boundaries.
+
+It deliberately avoids silently modifying Python environments, stopping unknown processes on port conflicts, writing recalled tool output back as conversation source, or persisting credentials.
+
+## TypeScript, OpenClaw, and DeepSeek Harness
+
+See [TypeScript Agent Integrations](./integrations/typescript.md) for the shared HTTP client, DeepSeek Harness adapter, OpenClaw adapter, and package validation.
+
+## Claude Code
+
+`integrations/claude_code/` provides streamable HTTP MCP configuration, a `reme-memory` Skill, and a Stop hook that calls `auto_memory_cc`. Follow that directory's README for installation.
+
+## Hermes Agent
+
+`integrations/hermes_agent/` provides a memory provider that recalls context before model calls and asynchronously invokes `auto_memory` after each turn.
+
+## Production guidance
+
+- choose a stable absolute `workspace_dir`;
+- reuse a service discovered by `reme find_reme`;
+- treat `reme help` as the active Job contract;
+- apply timeouts and failure logging to writes;
+- do not block the host's core response path when memory is temporarily unavailable;
+- use authentication, TLS, and a minimal Job allowlist for remote access.
--- a/docs/en/operations.md
+++ b/docs/en/operations.md
@ -0,0 +1,102 @@
+---
+title: Diagnostics, Backup, and Recovery
+description: ReMe health checks, logs, index maintenance, workspace backup, migration, and recovery.
+---
+
+# Diagnostics, Backup, and Recovery
+
+ReMe recovery protects user-owned workspace files and rebuilds catalogs, indexes, and graphs from those sources. Never delete or rewrite user memory merely to repair derived state.
+
+## Quick diagnosis
+
+Run these in order:
+
+```bash
+reme find_reme
+reme version
+reme health_check
+reme status
+reme app_config
+```
+
+- `find_reme` confirms the actual host, port, and PID;
+- `version` verifies that the CLI reaches the service;
+- `health_check` reports component health;
+- `status` estimates stateful component memory and process RSS;
+- `app_config` returns the effective configuration with secrets redacted.
+
+## Logs and common symptoms
+
+`log_to_console` and `log_to_file` control logging. For startup failures, inspect the first exception rather than later client connection errors.
+
+| Symptom | Check first |
+|---|---|
+| CLI cannot find ReMe | `reme find_reme`, process state, startup directory, and port |
+| Automatic memory fails | LLM backend, model, API key, and base URL |
+| Search is BM25-only | Whether an embedding store is connected to `file_store` |
+| New files are absent | Directory, extension, watcher, and `health_check` |
+| Studio fails but API works | Installed web extra, static path, and browser console |
+| Installed plugin is unavailable | Python interpreter, `plugins` configuration, and service restart |
+
+## Index maintenance
+
+```bash
+reme reindex scope=all
+reme reindex scope=bm25
+reme reindex scope=embedding
+```
+
+`reindex` rebuilds BM25 and/or embedding indexes from the current `file_chunks`. It does not scan the workspace, rechunk files, or rebuild the wikilink graph. Diagnose the watcher first when ingestion is the problem.
+
+Rebuild a daily index page separately:
+
+```bash
+reme daily_reindex date=2026-09-04
+```
+
+## Backup
+
+Stop writes or stop the service, then back up the complete workspace. The most important sources are:
+
+- `session/` for conversation sources;
+- `resource/` for external resources;
+- `daily/` for daily memory;
+- `digest/` for consolidated memory.
+
+`metadata/` contains indexes, graphs, and catalogs. Backing it up accelerates restoration, but it is not the sole source of truth.
+
+Use an explicit, stable absolute `workspace_dir` for durable deployments rather than relying on an incidental `.reme/` under the current directory.
+
+## Migrate a workspace
+
+1. Stop the old service to prevent writes during the copy.
+2. Copy the complete workspace while preserving timestamps.
+3. Start with the new absolute path:
+
+```bash
+reme start workspace_dir=/new/location/reme-memory
+```
+
+4. Run `health_check`, `status`, and a representative `search`.
+5. Rebuild embeddings if their model or dimensions changed.
+
+Do not push a workspace containing private conversations to a public repository.
+
+## Recover derived state
+
+Do not remove anything until a backup exists. Then:
+
+1. preserve `session/`, `resource/`, `daily/`, and `digest/`;
+2. record the effective configuration and component backends;
+3. verify that the failure is limited to `metadata/`;
+4. move suspect derived state to an isolated backup location;
+5. restart with the same configuration and let watchers rebuild;
+6. validate search, graph traversal, and daily indexes.
+
+The internal layout of metadata files is not a public automation contract.
+
+## Concurrent editing
+
+When Studio or an editor saves a complete file, pass the mtime from `stat` as `save.expected_mtime`. A save then fails if another actor changed the file after it was opened, avoiding silent overwrites.
+
+File Jobs enforce workspace containment and per-path locking. Do not bypass them to write arbitrary absolute paths.
--- a/docs/en/plugin_development.md
+++ b/docs/en/plugin_development.md
@ -0,0 +1,101 @@
+---
+title: Plugin Development
+description: Create, register, configure, test, and publish a ReMe plugin.
+---
+
+# Plugin Development
+
+A ReMe plugin is a regular Python distribution exposed through the `reme.plugins` entry-point group. Its package-level `plugin.yaml` can register Step and Component backends and provide default Application configuration.
+
+## Minimal structure
+
+```text
+my-plugin/
+├── pyproject.toml
+└── src/my_plugin/
+    ├── __init__.py
+    ├── plugin.yaml
+    └── steps.py
+```
+
+`pyproject.toml`:
+
+```toml
+[project.entry-points."reme.plugins"]
+my-plugin = "my_plugin"
+```
+
+`plugin.yaml`:
+
+```yaml
+name: my-plugin
+backends:
+  my_step: my_plugin.steps:MyStep
+application_defaults:
+  jobs:
+    my_action:
+      backend: base
+      description: Run my plugin action
+      parameters:
+        type: object
+        properties:
+          text: { type: string }
+        required: [text]
+      steps:
+        - backend: my_step
+```
+
+## Implement a Step
+
+```python
+from reme.components.component_registry import R
+from reme.steps.base_step import BaseStep
+
+
+@R.register("my_step")
+class MyStep(BaseStep):
+    async def execute(self):
+        self.context.response.answer = self.context.data["text"]
+```
+
+Step instances belong to one Job invocation. Put shared in-memory state under a narrow `app_context.metadata` key. Promote state that needs lifecycle, locking, or persistence to a Component or workspace file.
+
+## Configuration merge
+
+`application_defaults` is a partial `ApplicationConfig`:
+
+```text
+plugin defaults < selected/default config < CLI overrides
+```
+
+Plugins must not rewrite user configuration. Their backends enter an Application-local registry only when the plugin appears in that Application's `plugins` list.
+
+## Local validation
+
+```bash
+reme plugins validate ./path/to/my-plugin
+reme plugins install ./path/to/my-plugin --editable
+reme plugins list
+reme plugins show my-plugin
+reme start plugins='["my-plugin"]'
+reme my_action text=hello
+```
+
+Validation imports plugin code, so run it only for trusted sources.
+
+## Test boundaries
+
+- create workspaces with `tmp_path`;
+- mock network, model, and subprocess boundaries;
+- verify disabled plugins do not mutate the built-in registry;
+- verify plugin defaults and explicit configuration precedence;
+- keep tasks, clients, and executors under Component lifecycle;
+- never delete or rewrite user source files to repair derived state.
+
+The repository's Daily Paper, Auto Fin, LME, and BEAM plugins are complete examples.
+
+## Compatibility
+
+Legacy Python Plugin descriptors and the `reme.configs` entry point remain supported during migration, but new plugins should use `plugin.yaml`. Enablement always belongs to an Application rather than a process-global switch.
+
+See [Plugin Management](./plugin_management.md) for installation, upgrades, and removal.
--- a/docs/en/quick_start.md
+++ b/docs/en/quick_start.md
@ -1,5 +1,12 @@
+---
+title: Quick Start
+description: Install and start ReMe, then complete a first file, retrieval, and automatic-memory workflow.
+---
+
 # Quick Start

+This page gets one working loop running. See [Configuration](./configuration.md) for the full configuration contract and [Services and Deployment](./services.md) for HTTP or MCP integration.
+
 ## Installation

 ReMe requires Python 3.11+.
@ -237,3 +244,5 @@ You can also specify a YAML or JSON configuration file:
 ```bash
 reme start config=/path/to/custom.yaml
 ```
+
+Continue with the [CLI Reference](./reference/cli.md), [Job API Reference](./reference/jobs.md), or [Diagnostics, Backup, and Recovery](./operations.md).
--- a/docs/en/reference/cli.md
+++ b/docs/en/reference/cli.md
@ -0,0 +1,87 @@
+---
+title: CLI Reference
+description: ReMe command syntax, service invocation, configuration overrides, and plugin commands.
+---
+
+# CLI Reference
+
+The basic syntax is:
+
+```text
+reme ACTION key=value ...
+```
+
+## Start an Application
+
+```bash
+reme start
+reme start config=demo
+reme start workspace_dir=/data/reme service.port=8181
+reme start job=search query="keywords" limit=5
+```
+
+`start job=<name>` runs one Job through a one-shot service; plain `start` runs the configured Service.
+
+## Call Jobs
+
+Once a service is running, each action name is a Job name:
+
+```bash
+reme help
+reme health_check
+reme search query="project decision" limit=10
+reme read path=digest/wiki/project.md start_line=1 end_line=80
+```
+
+Use JSON for structured values:
+
+```bash
+reme auto_memory \
+  session_id=example \
+  messages='[{"role":"user","content":"Remember this preference"}]'
+```
+
+Client-selection arguments—`backend`, `transport`, `host`, `port`, `timeout`, `command`, `args`, and `show_metadata`—configure the client and never leak into the Job payload.
+
+## Configuration overrides
+
+```bash
+reme start \
+  config=/path/to/custom.yaml \
+  service.port=8181 \
+  service.web_enabled=false \
+  plugins='["auto-fin"]'
+```
+
+Leading `-` or `--` is optional. Use dots for nested keys and JSON for arrays and objects.
+
+## Service discovery
+
+```bash
+reme find_reme
+```
+
+This reports a discovered service but never starts or replaces a process.
+
+## Plugin commands
+
+Package management runs locally rather than through HTTP or MCP:
+
+```bash
+reme plugins list
+reme plugins show auto-fin
+reme plugins validate auto-fin
+reme plugins install reme-auto-fin
+reme plugins uninstall auto-fin
+```
+
+See [Plugin Management](../plugin_management.md) for the complete workflow.
+
+## Discover active capabilities
+
+The [Job API Reference](./jobs.md) describes the default configuration. Plugins and custom YAML may change the running service, so automation should prefer:
+
+```bash
+reme help
+reme app_config
+```
--- a/docs/en/services.md
+++ b/docs/en/services.md
@ -0,0 +1,115 @@
+---
+title: Services and Deployment
+description: Run ReMe through HTTP, SSE, MCP, the CLI, and ReMe Studio while respecting its local security boundary.
+---
+
+# Services and Deployment
+
+ReMe can run as a local HTTP service, a standalone MCP server, or a one-shot CLI Job. The default starts HTTP on `127.0.0.1:2333` and serves JSON, SSE, streamable HTTP MCP, and optional ReMe Studio from one process.
+
+## HTTP API
+
+```bash
+reme start
+reme start service.host=127.0.0.1 service.port=8181
+```
+
+Regular Jobs become `POST /<job-name>`:
+
+```bash
+curl -s http://127.0.0.1:2333/search \
+  -H 'Content-Type: application/json' \
+  -d '{"query":"user preferences","limit":5}'
+```
+
+Job arguments live at the request body's top level. A regular response follows the `Response` schema:
+
+```json
+{"success": true, "answer": "...", "metadata": {}}
+```
+
+Unhandled Step failures become unsuccessful responses.
+
+## Streaming and SSE
+
+Jobs with `backend: stream` also use `POST /<job-name>`, returning `text/event-stream`. Error paths emit an error chunk and always terminate the stream. The default `chat` Job is streaming:
+
+```bash
+curl -N http://127.0.0.1:2333/chat \
+  -H 'Content-Type: application/json' \
+  -d '{"query":"Summarize my long-term preferences"}'
+```
+
+MCP does not expose Stream Jobs.
+
+## MCP
+
+The default HTTP service mounts streamable HTTP MCP at `/mcp`:
+
+```yaml
+service:
+  backend: http
+  mcp_enabled: true
+  mcp_path: /mcp
+  mcp_stateless_http: false
+```
+
+For a standalone MCP service:
+
+```bash
+reme start service.backend=mcp service.transport=stdio
+reme start service.backend=mcp service.transport=sse service.port=2333
+reme start service.backend=mcp service.transport=streamable-http service.port=2333
+```
+
+MCP tools come from non-stream Jobs with `enable_serve: true`. Use `service.jobs` as an allowlist. `injected_job_kwargs` adds server-managed values that callers cannot override.
+
+```yaml
+service:
+  backend: http
+  jobs: [search, read, traverse, auto_memory]
+  injected_job_kwargs:
+    tenant_id: local-user
+  tool_error_on_failure: true
+```
+
+## ReMe Studio
+
+After installing `reme-ai[web]` or `reme-ai[core]`, the default HTTP origin also serves Studio:
+
+```text
+http://127.0.0.1:2333/
+```
+
+Disable it with `service.web_enabled=false` or select a build with `service.web_static_dir`. Missing static assets do not prevent the Job API from starting.
+
+## One-shot Jobs
+
+```bash
+reme start job=search query="user preferences" limit=5
+```
+
+This selects the one-shot CLI Service while retaining the normal Application, Component, and Job lifecycle.
+
+## Service discovery
+
+```bash
+reme find_reme
+```
+
+Ordinary `reme <action>` commands prefer the running service's actual backend, host, port, and transport. They fall back to local configuration only when no service is discovered.
+
+## Security boundary
+
+ReMe is local-first:
+
+- the default binds to `127.0.0.1`;
+- HTTP CORS allows any origin;
+- Jobs may write, move, or delete files;
+- the service layer has no general-purpose user authentication.
+
+Do not expose the default service directly to the public internet. For remote access, place it on a controlled network or behind an authenticated TLS reverse proxy, apply access controls and request-size limits, and expose only necessary Jobs.
+
+## OpenAPI
+
+FastAPI exposes the active endpoints through `/docs`, `/redoc`, and `/openapi.json`. The Studio SPA fallback preserves these reserved paths.
--- a/docs/index.md
+++ b/docs/index.md
@ -0,0 +1,43 @@
+---
+layout: home
+title: ReMe Documentation
+titleTemplate: false
+head:
+  - - meta
+    - http-equiv: refresh
+      content: "0; url=/zh/"
+  - - link
+    - rel: canonical
+      href: https://reme.agentscope.io/zh/
+hero:
+  name: ReMe
+  text: 让 Agent 真正记住
+  tagline: 文件属于你，记忆服务于 Agent。ReMe 将对话和资料沉淀为可读、可编辑、可检索、相互链接的本地文件。
+  image:
+    src: /favicon.svg
+    alt: ReMe
+  actions:
+    - theme: brand
+      text: 快速开始
+      link: /zh/quick_start
+    - theme: alt
+      text: 核心概念
+      link: /zh/memory_as_file
+features:
+  - icon: 📁
+    title: Local-first
+    details: Workspace 文件是持久事实源；索引、图谱和缓存都可以重新构建。
+    link: /zh/memory_as_file
+  - icon: 🧠
+    title: Memory workflows
+    details: 将对话和资料写入 daily，自动整理为互相关联的长期 digest。
+    link: /zh/auto_memory
+  - icon: 🔎
+    title: Search and graph
+    details: 结合关键词、可选向量检索与 wikilink 图谱，渐进式展开上下文。
+    link: /zh/memory_search
+  - icon: 🔌
+    title: Agent integrations
+    details: 通过 CLI、HTTP、MCP 和宿主适配器接入已有 Agent。
+    link: /zh/integrations
+---
--- a/docs/public/favicon.svg
+++ b/docs/public/favicon.svg
@ -0,0 +1,5 @@
+<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 64 64">
+  <defs><linearGradient id="g" x1="8" y1="7" x2="57" y2="58" gradientUnits="userSpaceOnUse"><stop stop-color="#19c9b0"/><stop offset="1" stop-color="#3156d9"/></linearGradient></defs>
+  <rect x="4" y="4" width="56" height="56" rx="18" fill="url(#g)"/>
+  <path d="M20 46V18h13.2c7.2 0 11.8 3.8 11.8 10 0 4.3-2.2 7.5-6 9l7 9h-8l-5.8-7.8H27V46h-7Zm7-13.6h5.6c3.5 0 5.4-1.4 5.4-4.2 0-2.8-1.9-4.1-5.4-4.1H27v8.3Z" fill="white"/>
+</svg>
--- a/docs/zh/configuration.md
+++ b/docs/zh/configuration.md
@ -0,0 +1,168 @@
+---
+title: 基础配置
+description: ReMe 配置文件、环境变量、命令行覆盖和核心组件配置。
+---
+
+# 基础配置
+
+ReMe 使用 YAML 或 JSON 描述 Service、Job 和 Component。默认配置位于 `reme/config/default.yaml`；启动时可以选择其他配置，再用命令行覆盖其中的字段。
+
+## 配置优先级
+
+配置按下面的顺序合并，靠后的值优先：
+
+1. 已启用插件提供的 `application_defaults`。
+2. 选中的配置文件；未指定时使用内置 `default`。
+3. 命令行 dot notation 覆盖。
+
+```bash
+reme start
+reme start config=demo
+reme start config=/absolute/path/to/app.yaml
+reme start service.port=8181 workspace_dir=/data/reme
+```
+
+`config` 支持内置配置名以及 `.yaml`、`.yml`、`.json` 文件。覆盖采用深度合并，不会因为修改 `service.port` 而丢失 `service` 下的其他字段。
+
+## 值的解析
+
+CLI 参数使用 `key=value`，前导 `-` 或 `--` 也会被接受：
+
+```bash
+reme start --service.port=8181 --service.web_enabled=false
+```
+
+值支持：
+
+- `null`、布尔值、整数和浮点数；
+- JSON 数组和对象；
+- JSON 引号字符串；
+- 普通字符串。
+
+类似 `007` 的前导零字符串不会被转换成数字。需要保留 `true`、`false` 等字面字符串时，使用 JSON 引号：`value='"true"'`。
+
+## 环境变量
+
+配置文件会递归展开两种表达式：
+
+```yaml
+api_key: ${LLM_API_KEY}
+base_url: ${LLM_BASE_URL:-https://example.com/v1}
+```
+
+`${VAR}` 在变量未定义时会报错；`${VAR:-default}` 使用默认值。ReMe 还会从命令启动目录向上查找 `.env`，最多检查五级父目录。
+
+不要把密钥提交到配置文件或 Git。推荐把密钥放在 `.env` 或进程环境中。
+
+## Application 字段
+
+| 字段 | 默认值 | 作用 |
+|---|---|---|
+| `app_name` | `ReMe` | 应用显示名称 |
+| `workspace_dir` | `.reme` | 用户拥有的 workspace 根目录，会规范化为绝对路径 |
+| `metadata_dir` | `metadata` | 索引、图谱和 catalog 等派生状态 |
+| `session_dir` | `session` | Agent 对话记录；标准 transcript 位于 `session/dialog` |
+| `mem_session_dir` | `mem_session` | Agent wrapper 的会话和配置 |
+| `resource_dir` | `resource` | 外部资料 |
+| `daily_dir` | `daily` | Daily memory |
+| `digest_dir` | `digest` | 长期整理后的记忆 |
+| `timezone` | `Asia/Shanghai` | Cron、日期和梦境流程使用的 IANA 时区 |
+| `language` | 空 | LLM 交互默认语言 |
+| `plugins` | `[]` | 为当前 Application 启用的已安装插件 |
+| `service` | HTTP | 服务端配置 |
+| `jobs` | 默认 Job | Job 名到 Job 配置的映射 |
+| `components` | 默认组件 | 按类型和名称组织的组件配置 |
+
+`session_dir` 必须保持 workspace-relative。其他 workspace 子目录也应使用清晰、稳定的相对名称。
+
+## LLM 配置
+
+默认 LLM 使用 OpenAI-compatible 接口：
+
+```yaml
+components:
+  as_llm:
+    default:
+      backend: openai
+      model: qwen3.7-plus
+      context_size: 200000
+      credential:
+        api_key: ${LLM_API_KEY:-}
+        base_url: ${LLM_BASE_URL:-}
+```
+
+可注册的内置 backend 包括 `openai`、`anthropic`、`dashscope`、`deepseek`、`gemini`、`moonshot`、`ollama` 和 `xai`。实际字段由对应 AgentScope model wrapper 决定。
+
+基础文件操作、BM25 检索、wikilink 遍历不需要 LLM。`auto_memory`、`auto_resource`、`auto_dream` 等演化流程需要可用 LLM。
+
+## Embedding 配置
+
+向量检索默认关闭。只设置 `EMBEDDING_API_KEY` 不会自动启用它；还需要同时启用 `as_embedding`、`embedding_store`，并把它连接到 `file_store`：
+
+```yaml
+components:
+  as_embedding:
+    default:
+      backend: openai
+      model: text-embedding-v4
+      dimensions: 1024
+      credential:
+        api_key: ${EMBEDDING_API_KEY}
+        base_url: ${EMBEDDING_BASE_URL:-https://dashscope.aliyuncs.com/compatible-mode/v1}
+  embedding_store:
+    default:
+      backend: local
+      as_embedding: default
+  file_store:
+    default:
+      backend: local
+      embedding_store: default
+      keyword_index: default
+      file_graph: default
+```
+
+修改 embedding 模型或维度后，应重新构建 embedding 索引。
+
+## Service 和 Job
+
+最小 HTTP 配置：
+
+```yaml
+service:
+  backend: http
+  host: 127.0.0.1
+  port: 2333
+  web_enabled: true
+  mcp_enabled: true
+  mcp_path: /mcp
+```
+
+Job 由 backend、参数 schema 和顺序执行的 Step 组成：
+
+```yaml
+jobs:
+  example:
+    backend: base
+    description: Example job
+    parameters:
+      type: object
+      properties:
+        text: { type: string }
+      required: [text]
+    steps:
+      - backend: example_step
+```
+
+设置 `enable_serve: false` 可以保留内部 Job、禁止 Service 暴露。后台和 Cron Job始终不会作为请求端点暴露。
+
+## 查看生效配置
+
+服务启动后运行：
+
+```bash
+reme app_config
+```
+
+返回的是已合并、已校验并隐藏密钥后的配置。排查覆盖顺序或插件配置时，应以它为准，而不是只查看某一个 YAML 文件。
+
+完整字段定义以 `reme/schema/application_config.py` 和 `reme/config/default.yaml` 为准。
--- a/docs/zh/contributing.md
+++ b/docs/zh/contributing.md
@ -190,6 +190,12 @@ pytest tests/unit/test_reme_cli.py
 docs/
 ```

+用户指南应同时提供 `docs/zh/` 与 `docs/en/` 版本，并在 `docs/.vitepress/config.mts` 的对应导航中注册。ReMe Studio、
+TypeScript、插件和评测的 README 是各自目录中的规范源文件；`github-pages/scripts/generate-content.mjs` 会在构建时镜像它们，
+不要编辑 `.generated/` 或 `dist/`。
+
+`Job API 参考`由 `reme/config/default.yaml` 自动生成。修改默认 Job 参数时更新 YAML 和测试，不要手工维护生成页。
+
 建议文档保持：

 - 标题明确，直接说明能力或流程。
@ -197,6 +203,15 @@ docs/
 - 涉及路径时使用仓库内真实路径，例如 `reme/config/default.yaml`、`reme/steps/`、`tests/unit/`。
 - 涉及默认行为时，以当前代码和 `pyproject.toml`、默认配置为准。

+文档站检查：
+
+```bash
+cd github-pages
+npm ci
+npm test
+npm run build
+```
+
 ---

 ## 获取帮助
--- a/docs/zh/faq.md
+++ b/docs/zh/faq.md
@ -0,0 +1,75 @@
+---
+title: 常见问题
+description: ReMe 安装、服务、模型、检索、文件和插件问题的快速答案。
+---
+
+# 常见问题
+
+## 基础文件操作需要模型 API Key 吗？
+
+不需要。`write`、`read`、`list`、`stat`、BM25 搜索和 wikilink 遍历可以在没有模型凭据时运行。`auto_memory`、`auto_resource` 和 `auto_dream` 需要 LLM。
+
+## 为什么配置了 Embedding Key 仍然只有 BM25？
+
+Embedding 默认未启用。除了凭据，还必须配置 `as_embedding`、`embedding_store`，并让 `file_store.default.embedding_store` 指向该组件。参见[基础配置](./configuration.md#embedding-配置)。
+
+## 为什么 `reme reindex` 没有发现新文件？
+
+`reindex` 只从当前 `file_chunks` 重建 BM25 或 embedding 索引，不扫描 workspace。检查 `index_update_loop` watcher、文件目录、后缀和 `health_check`。
+
+## 怎样使用另一个 workspace？
+
+启动时传入稳定的绝对路径：
+
+```bash
+reme start workspace_dir=/absolute/path/to/memory
+```
+
+普通 CLI 调用会发现运行服务，不必重复 workspace 参数。
+
+## 端口 2333 被占用怎么办？
+
+不要停止未知监听者。选择另一个端口：
+
+```bash
+reme start service.port=8181
+```
+
+然后用 `reme find_reme` 确认发现结果。
+
+## 为什么插件安装后仍然没有对应 Job？
+
+安装只让 distribution 在当前 Python 环境中可见。还需要在应用配置中启用：
+
+```bash
+reme start plugins='["auto-fin"]'
+```
+
+修改安装或启用状态后需要重启已运行的服务。
+
+## 可以直接编辑 workspace 中的 Markdown 吗？
+
+可以。文件是事实源，watcher 会摄取修改。应保留有效 frontmatter、使用完整 workspace-relative wikilink，并避免同时由多个编辑器无条件覆盖同一文件。
+
+## 可以把服务暴露到公网吗？
+
+默认配置不适合直接公网暴露。服务包含写入和删除 Job，HTTP CORS 宽松，且没有通用认证层。请使用受控网络或带 TLS、认证、访问控制的反向代理，并限制 `service.jobs`。
+
+## 怎样备份和迁移？
+
+停止写入后备份整个 workspace。`session/`、`resource/`、`daily/` 和 `digest/` 是最重要的用户数据；`metadata/` 可以随同备份，也可以从源文件重建。详见[诊断、备份与恢复](./operations.md)。
+
+## Studio 找不到怎么办？
+
+基础 `reme-ai` 包不包含前端资源。安装 `reme-ai[web]` 或 `reme-ai[core]`，或通过 `service.web_static_dir` 指向构建产物。Studio 缺失不会影响 Job API。
+
+## 当前服务到底开放了哪些能力？
+
+运行：
+
+```bash
+reme help
+reme app_config
+```
+
+静态文档描述默认配置；运行服务可能由自定义配置和插件改变。
--- a/docs/zh/index.md
+++ b/docs/zh/index.md
@ -0,0 +1,36 @@
+---
+layout: home
+title: ReMe 文档
+titleTemplate: false
+hero:
+  name: ReMe
+  text: 让 Agent 真正记住
+  tagline: 文件属于你，记忆服务于 Agent。ReMe 将对话和资料沉淀为可读、可编辑、可检索、相互链接的本地文件。
+  image:
+    src: /favicon.svg
+    alt: ReMe
+  actions:
+    - theme: brand
+      text: 快速开始
+      link: /zh/quick_start
+    - theme: alt
+      text: 查看配置
+      link: /zh/configuration
+features:
+  - icon: 📁
+    title: 文件即记忆
+    details: Workspace 文件是持久事实源；索引、图谱和缓存都可以重新构建。
+    link: /zh/memory_as_file
+  - icon: 🧠
+    title: 自动记忆
+    details: 将对话和资料写入 daily，自动整理为互相关联的长期 digest。
+    link: /zh/auto_memory
+  - icon: 🔎
+    title: 检索与图谱
+    details: 结合关键词、可选向量检索与 wikilink 图谱，渐进式展开上下文。
+    link: /zh/memory_search
+  - icon: 🔌
+    title: Agent 集成
+    details: 通过 CLI、HTTP、MCP 和宿主适配器接入已有 Agent。
+    link: /zh/integrations
+---
--- a/docs/zh/integrations.md
+++ b/docs/zh/integrations.md
@ -0,0 +1,88 @@
+---
+title: Agent 集成
+description: 通过 CLI、HTTP、MCP、Skill 和宿主适配器把 ReMe 接入 Agent。
+---
+
+# Agent 集成
+
+ReMe 把记忆能力放在独立服务和用户拥有的 workspace 中。Agent 可以通过标准接口调用同一套记忆，而不必把存储逻辑绑定到某一个模型或宿主。
+
+## 选择接入方式
+
+| 场景 | 推荐方式 |
+|---|---|
+| 本机脚本或 Hook | ReMe CLI |
+| 应用后端 | HTTP Client |
+| 支持工具协议的 Agent | MCP |
+| TypeScript Agent | `@agentscope-ai/reme` |
+| Claude Code | MCP + Skill + Stop Hook |
+| Hermes Agent | Memory provider adapter |
+| Codex 或其他 coding agent | `reme_memory` Skill 或 MCP |
+
+## 通用接入循环
+
+一个完整但可控的 Agent 记忆循环通常包含：
+
+1. 会话开始或回答前，用 `search` 找到相关记忆；
+2. 对高价值结果使用 `read`，必要时用 `traverse` 展开关系；
+3. 在回答中保留 workspace-relative 来源路径；
+4. 会话结束后，把原始消息交给 `auto_memory`；
+5. 由后台或定时任务把 daily 内容整理到 digest。
+
+搜索不到内容时应明确返回空结果，不应把模型推测当成历史记忆。
+
+## MCP
+
+默认 HTTP 服务在 `http://127.0.0.1:2333/mcp` 提供 streamable HTTP MCP。常用工具包括：
+
+- `search`
+- `read`
+- `traverse`
+- `list`
+- `auto_memory`
+- `proactive`
+
+根据宿主风险模型，可以用 `service.jobs` 只暴露只读工具，或将写入工具放在单独配置中。
+
+## CLI 和 Skill
+
+仓库中的 `skills/reme_memory/SKILL.md` 描述了一个通用 Agent 工作流，包括安装检查、服务发现、检索、读取和写入边界。它适合能够执行本地命令的 Agent。
+
+Skill 不应：
+
+- 未经允许安装或升级 Python 环境；
+- 发现端口冲突后停止未知进程；
+- 把召回的工具结果再次写入对话来源；
+- 将密钥或敏感信息写入记忆。
+
+## TypeScript、OpenClaw 与 DeepSeek Harness
+
+统一 TypeScript 客户端和宿主适配器见 [TypeScript Agent 集成](./integrations/typescript.md)。它包含：
+
+- HTTP Client；
+- DeepSeek Harness adapter；
+- OpenClaw adapter；
+- 构建与发布检查。
+
+## Claude Code
+
+仓库的 `integrations/claude_code/` 提供：
+
+- streamable HTTP MCP 配置；
+- `reme-memory` Skill；
+- 会话停止时调用 `auto_memory_cc` 的 Hook。
+
+完整安装步骤以仓库中的 `integrations/claude_code/README.md` 为准。
+
+## Hermes Agent
+
+`integrations/hermes_agent/` 提供 memory provider：模型调用前检索相关记忆，每轮结束后异步调用 `auto_memory`。完整配置见该目录 README。
+
+## 生产接入建议
+
+- 明确选择稳定、绝对的 `workspace_dir`；
+- 启动前复用 `reme find_reme` 发现的服务；
+- 以 `reme help` 为当前 Job 契约；
+- 为写入动作设置超时和失败日志；
+- 不因记忆服务暂时不可用而阻塞宿主的核心回答流程；
+- 对远程访问使用认证、TLS 和最小 Job allowlist。
--- a/docs/zh/integrations/claude-code.md
+++ b/docs/zh/integrations/claude-code.md
@ -0,0 +1,67 @@
+---
+title: Claude Code 集成
+description: 通过 MCP、reme-memory Skill 和 Stop Hook 将 Claude Code 连接到 ReMe。
+---
+
+# Claude Code 集成
+
+ReMe 的 Claude Code 插件提供长期记忆召回，并在每次会话结束后异步记录对话。Daily 到 digest 的整理仍由共享的 ReMe 服务负责。
+
+## 能力
+
+- 通过 MCP 使用 `search`、`traverse`、`daily_list`、`frontmatter_read`、`read`、`auto_memory_cc` 等工具；
+- `reme-memory` Skill 在回答前召回长期记忆并保留来源路径；
+- Stop Hook 只把 Claude Code `session_id` 传给服务端，服务端从本地 transcript 解析会话；
+- 记录在脱离 Claude Code 的后台进程中进行，不延迟退出；服务不可用时记录日志并结束。
+
+## 部署模型
+
+插件连接到用户预先启动的共享 HTTP MCP 服务，不为每个 Claude Code 窗口创建 ReMe。这样所有窗口共享一个 workspace、一组 watcher 和一次 dream cron。
+
+## 准备 ReMe
+
+```bash
+pip install "reme-ai[core]"
+```
+
+在稳定目录配置 LLM 环境，然后启动：
+
+```bash
+reme start service.backend=http
+```
+
+默认 JSON Job API 和 MCP 地址分别位于同一个 `127.0.0.1:2333` 服务，MCP 路径是 `/mcp`。使用其他端口时，必须同步修改插件的 `.mcp.json`。
+
+默认搜索使用 BM25；只有启用向量检索时才需要 Embedding 配置。
+
+## 安装插件
+
+在 Claude Code 中运行：
+
+```text
+/plugin marketplace add ./integrations/claude_code
+/plugin install reme@reme-marketplace
+```
+
+重启 Claude Code，再运行 `/mcp`，确认 `reme` server 和工具已经连接。
+
+## Hook 与路径
+
+- MCP 配置：`integrations/claude_code/reme/.mcp.json`；
+- 自动记忆 Hook：`integrations/claude_code/reme/hooks/auto_memory.py`；
+- Hook 日志：`integrations/claude_code/reme/logs/auto_memory_hook.log`；
+- 默认 transcript 根目录：`~/.claude/projects`；
+- 可通过 `CLAUDE_CONFIG_DIR` 修改 transcript 根目录；
+- 可通过 `REME_HOST`、`REME_PORT` 覆盖 Hook 使用的服务地址。
+
+Hook 需要 `python3` 位于 `PATH`。MCP 工具名前缀可能随 Claude Code 版本包含 server segment；Skill 使用 `mcp__reme__*` 匹配这一差异。
+
+## 验证
+
+1. `reme health_check` 返回健康；
+2. Claude Code `/mcp` 显示 ReMe；
+3. `reme-memory` 能召回一条已存在记忆；
+4. 完成测试会话后，Hook 日志没有错误；
+5. 对应内容出现在当天 daily note 中。
+
+英文原始部署说明位于 `integrations/claude_code/README.md`。
--- a/docs/zh/integrations/hermes.md
+++ b/docs/zh/integrations/hermes.md
@ -0,0 +1,59 @@
+---
+title: Hermes Agent 集成
+description: 使用 ReMe memory provider 在 Hermes 调用模型前召回、每轮结束后异步记录。
+---
+
+# Hermes Agent 集成
+
+Hermes memory provider 连接到一个已经运行的 ReMe HTTP 服务，在每次模型调用前召回相关记忆，并在用户/助手回合完成后异步调用 `auto_memory`。
+
+## Workspace 隔离
+
+ReMe 的搜索范围是一个完整 workspace。多个 Hermes profile 指向同一个 workspace 时会共享召回结果；需要隔离时，为每个 profile 使用独立 workspace 和端点。
+
+```bash
+reme start \
+  workspace_dir="/absolute/path/to/reme-hermes-default" \
+  service.backend=http \
+  service.host=127.0.0.1 \
+  service.port=2333
+```
+
+自动记忆需要 LLM；默认 BM25 搜索不需要 Embedding。
+
+## 安装与配置
+
+```bash
+hermes plugins install agentscope-ai/ReMe/integrations/hermes_agent
+hermes memory setup
+```
+
+选择 `reme`，接受默认的 `http://127.0.0.1:2333`，或输入上一步使用的端点。Setup 会先调用 `health_check`，只有新端点健康时才替换现有 provider 配置。
+
+配置存放在 `$HERMES_HOME/reme.json`：
+
+```json
+{
+  "endpoint": "http://127.0.0.1:2333",
+  "request_timeout": 600.0,
+  "recall_timeout": 5.0,
+  "health_timeout": 2.0,
+  "health_retry_seconds": 30.0,
+  "shutdown_timeout": 30.0,
+  "recall_limit": 5
+}
+```
+
+运行 `hermes memory status` 检查安装和配置。新的 Hermes 会话还会重新检查端点健康状态。
+
+## 生命周期和失败行为
+
+- `prefetch` 调用 ReMe `search`，Hermes 将结果放入受保护的 memory context；
+- `sync_turn` 把完成的回合加入串行后台写队列，再调用 `auto_memory`；
+- cron、flush 和 subagent context 不写入对话记忆；
+- 健康检查失败后，在 cooldown 结束前暂停召回和记录；
+- 召回与写入有独立 cooldown，单项失败不会关闭另一项；
+- 召回使用较短超时，避免慢搜索长期阻塞模型调用；
+- shutdown 会在有限时间内排空写队列，ReMe 服务仍由用户独立管理。
+
+英文权威安装说明位于 `integrations/hermes_agent/README.md`。
--- a/docs/zh/operations.md
+++ b/docs/zh/operations.md
@ -0,0 +1,106 @@
+---
+title: 诊断、备份与恢复
+description: ReMe 服务健康检查、日志、索引维护、备份迁移和常见恢复流程。
+---
+
+# 诊断、备份与恢复
+
+ReMe 的恢复原则是：保护 workspace 中的用户文件，通过源文件重建索引、图谱和 catalog。不要为了修复索引而删除或改写用户记忆。
+
+## 快速诊断
+
+依次执行：
+
+```bash
+reme find_reme
+reme version
+reme health_check
+reme status
+reme app_config
+```
+
+- `find_reme`：服务是否存在，以及实际 host、port、PID；
+- `version`：CLI 能否成功访问服务；
+- `health_check`：组件健康状态；
+- `status`：状态组件的内存估算和进程 RSS；
+- `app_config`：隐藏密钥后的实际生效配置。
+
+## 日志
+
+`log_to_console` 和 `log_to_file` 控制日志目标。排查启动失败时先查看第一条异常，而不是后续客户端连接错误。
+
+常见类别：
+
+| 现象 | 优先检查 |
+|---|---|
+| CLI 找不到服务 | `reme find_reme`、启动目录、端口和进程状态 |
+| 自动记忆失败 | LLM backend、model、API key、base URL |
+| 只有 BM25 结果 | embedding 组件是否真正接入 `file_store` |
+| 新文件没有进入搜索 | 文件所在目录、后缀、watcher 和 `health_check` |
+| Studio 空白但 API 正常 | web extra、静态构建路径和浏览器控制台 |
+| 插件安装后不可用 | 当前 Python 解释器、`plugins` 配置、服务重启 |
+
+## 索引维护
+
+```bash
+reme reindex scope=all
+reme reindex scope=bm25
+reme reindex scope=embedding
+```
+
+`reindex` 从当前 `file_chunks` 重建 BM25 和/或 embedding 派生索引。它不会扫描 workspace、重新分块或重建 wikilink 图谱。
+
+如果问题发生在文件摄取阶段，应先确认后台 watcher 正常运行；不能把 `reindex` 当成通用“重新扫描”命令。
+
+Daily 索引页可单独重建：
+
+```bash
+reme daily_reindex date=2026-09-04
+```
+
+## 备份
+
+停止写入或停止服务后，优先备份整个 workspace。最重要的目录是：
+
+- `session/`：原始对话来源；
+- `resource/`：外部资料；
+- `daily/`：每日记忆；
+- `digest/`：长期记忆。
+
+`metadata/` 包含索引、图谱和 file catalog。一起备份可以加速恢复，但它不是唯一事实源。
+
+不要只备份进程目录下偶然生成的 `.reme/`；生产使用应明确设置稳定的绝对 `workspace_dir`。
+
+## 迁移 workspace
+
+1. 停止旧服务，避免迁移期间继续写入。
+2. 复制完整 workspace，并保留文件时间信息。
+3. 使用新的绝对路径启动：
+
+```bash
+reme start workspace_dir=/new/location/reme-memory
+```
+
+4. 运行 `health_check`、`status` 和一次代表性 `search`。
+5. 如果 embedding 模型或维度发生变化，再重建 embedding 索引。
+
+普通 Markdown 和资源文件可以使用版本控制或同步工具；包含敏感对话的 workspace 不应推送到公开仓库。
+
+## 从派生状态故障恢复
+
+在确认备份可用前，不要删除任何内容。恢复顺序应为：
+
+1. 保留 `session/`、`resource/`、`daily/` 和 `digest/`；
+2. 记录当前配置和组件 backend；
+3. 确认故障只发生在 `metadata/`；
+4. 将有问题的派生状态移到隔离备份位置；
+5. 用同一配置启动 ReMe，让 watcher 从源文件重建；
+6. 验证搜索、图谱和 Daily 索引。
+
+具体 metadata 文件属于实现细节，不应在自动化脚本里依赖其内部格式。
+
+## 并发编辑
+
+Studio 或其他编辑器保存完整文件时，应使用 `stat` 返回的 mtime 作为 `save.expected_mtime`。如果文件在打开后被外部修改，保存会失败，从而避免静默覆盖。
+
+文件 Job 会校验 workspace containment，并对同一路径加锁。不要绕过这些 Job 直接向不受控制的绝对路径写入。
--- a/docs/zh/plugin_development.md
+++ b/docs/zh/plugin_development.md
@ -0,0 +1,101 @@
+---
+title: 插件开发
+description: 创建、注册、配置、测试和发布 ReMe 插件。
+---
+
+# 插件开发
+
+ReMe 插件是一个普通 Python distribution，通过 `reme.plugins` entry-point group 暴露 package-level `plugin.yaml`。插件可以注册新的 Step、Component backend，并提供默认 Application 配置。
+
+## 最小结构
+
+```text
+my-plugin/
+├── pyproject.toml
+└── src/my_plugin/
+    ├── __init__.py
+    ├── plugin.yaml
+    └── steps.py
+```
+
+`pyproject.toml`：
+
+```toml
+[project.entry-points."reme.plugins"]
+my-plugin = "my_plugin"
+```
+
+`plugin.yaml`：
+
+```yaml
+name: my-plugin
+backends:
+  my_step: my_plugin.steps:MyStep
+application_defaults:
+  jobs:
+    my_action:
+      backend: base
+      description: Run my plugin action
+      parameters:
+        type: object
+        properties:
+          text: { type: string }
+        required: [text]
+      steps:
+        - backend: my_step
+```
+
+## 实现 Step
+
+```python
+from reme.components.component_registry import R
+from reme.steps.base_step import BaseStep
+
+
+@R.register("my_step")
+class MyStep(BaseStep):
+    async def execute(self):
+        self.context.response.answer = self.context.data["text"]
+```
+
+Step 实例属于单次 Job 调用。跨调用的内存状态应放在带命名空间的 `app_context.metadata`；需要生命周期、锁或持久化时，应升级为 Component 或 workspace 文件。
+
+## 配置合并
+
+`application_defaults` 是不完整的 `ApplicationConfig`。合并顺序为：
+
+```text
+插件默认值 < 选中/default 配置 < CLI 覆盖
+```
+
+插件不应自动修改用户的配置文件。只有在 Application 的 `plugins` 列表中显式启用后，插件 backend 才会加入该 Application 的局部 registry。
+
+## 本地验证
+
+```bash
+reme plugins validate ./path/to/my-plugin
+reme plugins install ./path/to/my-plugin --editable
+reme plugins list
+reme plugins show my-plugin
+reme start plugins='["my-plugin"]'
+reme my_action text=hello
+```
+
+校验会导入插件代码，因此只应对可信源码执行。
+
+## 测试边界
+
+- 使用 `tmp_path` 创建 workspace；
+- Mock 网络、模型和子进程；
+- 验证插件未启用时不会污染 built-in registry；
+- 验证默认配置与显式配置的合并优先级；
+- 验证后台任务、客户端和 executor 都由 Component 生命周期关闭；
+- 验证失败不会删除或重写用户源文件。
+
+仓库中的 `plugins/daily_paper`、`plugins/auto-fin`、`plugins/lme` 和 `plugins/beam` 是完整参考实现。
+
+## 兼容性
+
+迁移期间仍兼容旧的 Python Plugin descriptor 和 `reme.configs` entry point，但新插件应使用 `plugin.yaml`。不要依赖进程级全局 enable/disable 状态；插件启用始终属于具体 Application 配置。
+
+包管理命令、升级与卸载行为见[插件管理](./plugin_management.md)。
--- a/docs/zh/quick_start.md
+++ b/docs/zh/quick_start.md
@ -1,5 +1,12 @@
+---
+title: 快速开始
+description: 安装并启动 ReMe，完成文件写入、检索和自动记忆的第一个闭环。
+---
+
 # 快速开始

+本页用于完成第一次可运行闭环。需要完整配置字段时查看[基础配置](./configuration.md)；接入 HTTP 或 MCP 时查看[服务与部署](./services.md)。
+
 ## 安装

 ReMe 要求 Python 3.11+。
@ -228,3 +235,5 @@ reme start \
 ```bash
 reme start config=/path/to/custom.yaml
 ```
+
+下一步可以查看 [CLI 参考](./reference/cli.md)、[Job API 参考](./reference/jobs.md)和[诊断、备份与恢复](./operations.md)。
--- a/docs/zh/reference/cli.md
+++ b/docs/zh/reference/cli.md
@ -0,0 +1,87 @@
+---
+title: CLI 参考
+description: ReMe 命令行语法、服务调用、配置覆盖与插件命令。
+---
+
+# CLI 参考
+
+ReMe 的基本语法是：
+
+```text
+reme ACTION key=value ...
+```
+
+## 启动应用
+
+```bash
+reme start
+reme start config=demo
+reme start workspace_dir=/data/reme service.port=8181
+reme start job=search query="关键词" limit=5
+```
+
+`start job=<name>` 运行一次性 Job；普通 `start` 启动配置中的 Service。
+
+## 调用 Job
+
+服务运行后，Action 名就是 Job 名：
+
+```bash
+reme help
+reme health_check
+reme search query="项目决策" limit=10
+reme read path=digest/wiki/project.md start_line=1 end_line=80
+```
+
+复杂值使用 JSON：
+
+```bash
+reme auto_memory \
+  session_id=example \
+  messages='[{"role":"user","content":"记住这条偏好"}]'
+```
+
+服务选择参数 `backend`、`transport`、`host`、`port`、`timeout`、`command`、`args` 和 `show_metadata` 只用于构造客户端，不会泄漏到 Job 参数。
+
+## 配置覆盖
+
+```bash
+reme start \
+  config=/path/to/custom.yaml \
+  service.port=8181 \
+  service.web_enabled=false \
+  plugins='["auto-fin"]'
+```
+
+参数前的 `-` 或 `--` 可以省略。嵌套键使用点号，数组和对象使用 JSON。
+
+## 服务发现
+
+```bash
+reme find_reme
+```
+
+成功时输出可复用的服务信息；找不到服务时不会自动启动新进程。
+
+## 插件命令
+
+插件包管理是本地命令，不经过 HTTP 或 MCP：
+
+```bash
+reme plugins list
+reme plugins show auto-fin
+reme plugins validate auto-fin
+reme plugins install reme-auto-fin
+reme plugins uninstall auto-fin
+```
+
+完整说明见[插件管理](../plugin_management.md)。
+
+## 发现当前能力
+
+默认 Job 参数见 [Job API 参考](./jobs.md)。运行配置可能由插件和自定义 YAML 改变，因此自动化程序应优先调用：
+
+```bash
+reme help
+reme app_config
+```
--- a/docs/zh/services.md
+++ b/docs/zh/services.md
@ -0,0 +1,131 @@
+---
+title: 服务与部署
+description: 使用 ReMe 的 HTTP、SSE、MCP 和 Studio 服务，并理解默认安全边界。
+---
+
+# 服务与部署
+
+ReMe 可以作为本地 HTTP 服务、独立 MCP Server 或一次性 CLI Job 运行。默认模式是在 `127.0.0.1:2333` 启动 HTTP 服务，并在同一进程中提供 JSON API、SSE、MCP 与可选的 ReMe Studio。
+
+## HTTP 服务
+
+```bash
+reme start
+reme start service.host=127.0.0.1 service.port=8181
+```
+
+普通 Job 暴露为 `POST /<job-name>`：
+
+```bash
+curl -s http://127.0.0.1:2333/search \
+  -H 'Content-Type: application/json' \
+  -d '{"query":"用户偏好","limit":5}'
+```
+
+请求体允许 Job 参数位于顶层。响应遵循 `Response`：
+
+```json
+{
+  "success": true,
+  "answer": "...",
+  "metadata": {}
+}
+```
+
+未被 Step 捕获的错误会转换成 `success: false` 的响应。
+
+## Streaming 与 SSE
+
+`backend: stream` 的 Job 仍使用 `POST /<job-name>`，但响应类型是 `text/event-stream`。每个 chunk 使用统一的 streaming schema；失败时会发出错误 chunk，并以终止事件结束。
+
+默认 `chat` 是 Stream Job：
+
+```bash
+curl -N http://127.0.0.1:2333/chat \
+  -H 'Content-Type: application/json' \
+  -d '{"query":"总结我的长期偏好"}'
+```
+
+MCP 不暴露 Stream Job。
+
+## MCP
+
+默认 HTTP 服务会在 `/mcp` 挂载 streamable HTTP MCP：
+
+```yaml
+service:
+  backend: http
+  mcp_enabled: true
+  mcp_path: /mcp
+  mcp_stateless_http: false
+```
+
+也可以改用独立 MCP Service：
+
+```bash
+reme start service.backend=mcp service.transport=stdio
+reme start service.backend=mcp service.transport=sse service.port=2333
+reme start service.backend=mcp service.transport=streamable-http service.port=2333
+```
+
+MCP tool 来自 `enable_serve: true` 的非流式 Job。`service.jobs` 可以设置允许列表；`injected_job_kwargs` 可以注入服务端管理、调用方不能覆盖的参数。
+
+```yaml
+service:
+  backend: http
+  jobs: [search, read, traverse, auto_memory]
+  injected_job_kwargs:
+    tenant_id: local-user
+  tool_error_on_failure: true
+```
+
+## ReMe Studio
+
+安装 `reme-ai[web]` 或 `reme-ai[core]` 后，默认 HTTP 地址同时提供 ReMe Studio：
+
+```text
+http://127.0.0.1:2333/
+```
+
+可通过配置关闭或指定自定义构建：
+
+```yaml
+service:
+  web_enabled: false
+  # web_static_dir: /absolute/path/to/static
+```
+
+静态资源缺失不会阻止 Job API 启动。
+
+## 一次性 Job
+
+需要脚本式执行而不常驻服务时：
+
+```bash
+reme start job=search query="用户偏好" limit=5
+```
+
+它会切换到一次性 CLI Service，但仍经过正常的 Application、Component 和 Job 生命周期。
+
+## 服务发现
+
+```bash
+reme find_reme
+```
+
+ReMe 会记录本机运行服务的启动参数。普通 `reme <action>` 优先使用实际运行服务的 backend、host、port 和 transport；找不到运行服务时才回退到本地配置。
+
+## 安全边界
+
+ReMe 默认定位为本地服务：
+
+- 默认绑定 `127.0.0.1`；
+- HTTP CORS 配置允许任意 origin；
+- Job 可进行文件写入、移动和删除；
+- 当前服务层不提供通用用户认证。
+
+不要直接把默认服务暴露到公网。需要远程访问时，在受控网络或带身份认证、TLS、访问控制和请求大小限制的反向代理后部署，并通过 `service.jobs` 只开放必要 Job。
+
+## OpenAPI
+
+HTTP 服务使用 FastAPI。启用 HTTP 服务时，可以通过 `/docs`、`/redoc` 和 `/openapi.json` 查看当前配置实际注册的端点；Studio 的 SPA fallback 不会覆盖这些保留路径。
--- a/github-pages/README.md
+++ b/github-pages/README.md
@ -1,8 +1,8 @@
-# ReMe GitHub Pages
+# ReMe documentation site

-This directory contains the standalone Vite documentation site published at <https://reme.agentscope.io>. The
-GitHub Pages fallback is <https://agentscope-ai.github.io/ReMe/>. It does not depend on the ReMe Studio application in
-`reme_studio/`.
+This package builds the VitePress site published at <https://reme.agentscope.io>. The canonical documentation lives in
+`docs/`; selected product, integration, plugin, and benchmark READMEs are mirrored into a disposable source tree during
+the build. Do not edit `.generated/` or `dist/`.

 ## Requirements

@ -11,79 +11,41 @@ GitHub Pages fallback is <https://agentscope-ai.github.io/ReMe/>. It does not de

 ## Local development

-From the repository root:
-
 ```bash
 cd github-pages
-npm install
+npm ci
 npm run dev
 ```

-Open the URL printed by Vite, normally <http://localhost:5173/>. The development server watches the frontend source.
-When a repository Markdown file changes, restart the development command to regenerate the documentation content.
+The development server prints its local URL. Restart it after changing a mirrored README or `reme/config/default.yaml`
+so the generated source tree and Job reference are refreshed. Changes under `docs/` are also refreshed on restart.

-For subsequent installs or CI-compatible dependency installation, use:
-
-```bash
-npm ci
-```
-
-## Preview the production build
-
-Build and start the preview server:
+## Validation

 ```bash
+npm test
 npm run build
 npm run preview
 ```

-Open the URL printed by Vite, normally <http://localhost:4173/>. Production assets use relative paths so the same build
-works on both the custom domain and the GitHub Pages project path.
+The test suite verifies bilingual core pages, canonical-source mappings, generated Job coverage, and disposable output.
+The production build is written to `github-pages/dist/` for the existing GitHub Pages workflow.

-The generated `dist/` and `.generated/` directories are disposable build output and are excluded from Git.
+## Sources

-## Documentation sources
-
-The build script reads the canonical repository files directly. Do not edit generated copies under `.generated/` or
-`dist/`.
-
- `README.md` and `README_ZH.md`: project introductions
- `docs/en/` and `docs/zh/`: English and Chinese guides
- `docs/figure/`: documentation images
- `reme_studio/README.md` and `reme_studio/README_ZH.md`: ReMe Studio guide
- `typescript/README.md` and `typescript/README_ZH.md`: TypeScript client, DeepSeek Harness, and OpenClaw integration guide
+- `docs/`: canonical guides, VitePress configuration, theme, and brand assets
+- `reme/config/default.yaml`: generated callable Job reference
+- `reme_studio/README*.md`: ReMe Studio
+- `typescript/README*.md`: TypeScript client and adapters
 - `plugins/*/README*.md`: plugin guides
- `benchmark/{beam,longmemeval,pibench,toolmemory}/README*.md`: benchmark guides and results
- `AGENTS.md`: repository development guide
+- `benchmark/*/README*.md`: benchmark guides
+- `scripts/generate-content.mjs`: source mirroring and reference generation

-To add or reorganize a document in the site navigation, update
-[`scripts/generate-content.mjs`](./scripts/generate-content.mjs). Presentation and interaction code lives in `src/`.
-
-## Project structure
-
-```text
-github-pages/
-├── index.html
-├── package.json
-├── scripts/
-│   └── generate-content.mjs
-├── src/
-│   ├── main.js
-│   └── styles.css
-└── vite.config.js
-```
+When adding a canonical guide, add both `docs/zh/<name>.md` and `docs/en/<name>.md`, then include it in the appropriate
+sidebar in `docs/.vitepress/config.mts`. Add repository-owned READMEs to `externalDocuments` in the generator rather than
+duplicating their full content under `docs/`.

 ## Deployment

-The repository workflow `.github/workflows/deploy-docs.yml` builds this directory and publishes `dist/` to GitHub Pages.
-It runs after relevant documentation or site files change on `main`, and it can also be started manually from the
-GitHub Actions page.
-
-The repository's **Settings → Pages → Build and deployment → Source** must be set to **GitHub Actions**. Its custom
-domain must be set to `reme.agentscope.io`; `public/CNAME` preserves that domain in the published artifact.
-
-Useful links:
-
- ReMe documentation: <https://reme.agentscope.io>
- GitHub Pages fallback: <https://agentscope-ai.github.io/ReMe/>
- ReMe repository: <https://github.com/agentscope-ai/ReMe>
+`.github/workflows/deploy-docs.yml` uses the reusable documentation build workflow and publishes `dist/` to GitHub
+Pages. `public/CNAME` preserves the `reme.agentscope.io` custom domain.
--- a/github-pages/index.html
+++ b/github-pages/index.html
@ -1,17 +0,0 @@
-<!doctype html>
-<html lang="zh-CN">
-  <head>
-    <meta charset="UTF-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <meta
-      name="description"
-      content="ReMe documentation — a local-first, file-native memory system for agents."
-    />
-    <link rel="icon" type="image/svg+xml" href="%BASE_URL%favicon.svg" />
-    <title>ReMe Documentation</title>
-  </head>
-  <body>
-    <div id="app"></div>
-    <script type="module" src="/src/main.js"></script>
-  </body>
-</html>
--- a/github-pages/package-lock.json
+++ b/github-pages/package-lock.json
--- a/github-pages/package.json
+++ b/github-pages/package.json
@ -7,16 +7,14 @@
    "node": ">=22.13.0"
  },
  "scripts": {
-    "dev": "node scripts/generate-content.mjs && vite",
-    "build": "node scripts/generate-content.mjs && vite build",
-    "preview": "vite preview",
-    "test": "node scripts/generate-content.mjs && node --test tests/*.test.mjs"
-  },
-  "dependencies": {
-    "dompurify": "^3.2.6",
-    "marked": "^16.2.1"
+    "content": "node scripts/generate-content.mjs",
+    "dev": "npm run content && vitepress dev .generated/site",
+    "build": "npm run content && vitepress build .generated/site --outDir dist && node scripts/verify-build.mjs",
+    "preview": "vitepress preview .generated/site --outDir dist",
+    "test": "npm run content && node --test tests/*.test.mjs"
  },
  "devDependencies": {
-    "vite": "^7.1.1"
+    "vitepress": "1.6.4",
+    "yaml": "^2.8.1"
  }
 }
--- a/github-pages/scripts/generate-content.mjs
+++ b/github-pages/scripts/generate-content.mjs
@ -1,249 +1,191 @@
 import { cp, mkdir, readFile, rm, writeFile } from "node:fs/promises";
 import path from "node:path";
 import { fileURLToPath } from "node:url";
+import { parse as parseYaml } from "yaml";

 const siteDir = path.resolve(path.dirname(fileURLToPath(import.meta.url)), "..");
 const repoDir = path.resolve(siteDir, "..");
-const outputDir = path.join(siteDir, ".generated", "content");
+const outputDir = path.join(siteDir, ".generated", "site");

-const navigationGroupOrder = [
-  "overview",
-  "start",
-  "integration",
-  "fundamentals",
-  "automation",
-  "concepts",
-  "workspace",
-  "plugins",
-  "benchmarks",
-  "development",
+const externalDocuments = [
+  ["en/integrations/claude-code.md", "integrations/claude_code/README.md"],
+  ["en/integrations/hermes.md", "integrations/hermes_agent/README.md"],
+  ["zh/integrations/typescript.md", "typescript/README_ZH.md"],
+  ["en/integrations/typescript.md", "typescript/README.md"],
+  ["zh/workspace/studio.md", "reme_studio/README_ZH.md"],
+  ["en/workspace/studio.md", "reme_studio/README.md"],
+  ["zh/plugins/daily-paper.md", "plugins/daily_paper/README_ZH.md"],
+  ["en/plugins/daily-paper.md", "plugins/daily_paper/README.md"],
+  ["zh/plugins/auto-fin.md", "plugins/auto-fin/README_ZH.md"],
+  ["en/plugins/auto-fin.md", "plugins/auto-fin/README.md"],
+  ["zh/plugins/lme.md", "plugins/lme/README_ZH.md"],
+  ["en/plugins/lme.md", "plugins/lme/README.md"],
+  ["zh/plugins/beam.md", "plugins/beam/README_ZH.md"],
+  ["en/plugins/beam.md", "plugins/beam/README.md"],
+  ["zh/benchmarks/beam.md", "benchmark/beam/README_ZH.md"],
+  ["en/benchmarks/beam.md", "benchmark/beam/README.md"],
+  ["zh/benchmarks/longmemeval.md", "benchmark/longmemeval/README_ZH.md"],
+  ["en/benchmarks/longmemeval.md", "benchmark/longmemeval/README.md"],
+  ["zh/benchmarks/pibench.md", "benchmark/pibench/README_ZH.md"],
+  ["en/benchmarks/pibench.md", "benchmark/pibench/README.md"],
+  ["zh/benchmarks/toolmemory.md", "benchmark/toolmemory/README_ZH.md"],
+  ["en/benchmarks/toolmemory.md", "benchmark/toolmemory/README.md"],
 ];

-const topicOrder = [
-  "quick_start",
-  "plugin_management",
-  "memory_as_file",
-  "memory_search",
-  "auto_memory",
-  "auto_resource",
-  "auto_link",
-  "auto_dream",
-  "proactive",
-  "reme_scene",
-  "framework",
-  "reme-blog",
-  "contributing",
-];
-
-const groups = {
-  quick_start: "start",
-  plugin_management: "start",
-  memory_as_file: "fundamentals",
-  memory_search: "fundamentals",
-  auto_memory: "automation",
-  auto_resource: "automation",
-  auto_link: "automation",
-  auto_dream: "automation",
-  proactive: "automation",
-  reme_scene: "concepts",
-  framework: "concepts",
-  "reme-blog": "concepts",
-  contributing: "development",
+const groupNames = {
+  zh: {
+    system: "系统与诊断",
+    memory: "记忆演化",
+    retrieval: "检索与图谱",
+    daily: "Daily Note",
+    files: "文件操作",
+  },
+  en: {
+    system: "System and diagnostics",
+    memory: "Memory evolution",
+    retrieval: "Retrieval and graph",
+    daily: "Daily notes",
+    files: "File operations",
+  },
 };

-const localizedTitles = {
-  quick_start: { zh: "快速开始", en: "Quick Start" },
-  plugin_management: { zh: "插件管理", en: "Plugin Management" },
-  memory_as_file: { zh: "文件即记忆", en: "Memory as File" },
-  memory_search: { zh: "记忆检索", en: "Memory Search" },
-  auto_memory: { zh: "自动记忆", en: "Auto Memory" },
-  auto_resource: { zh: "自动资料整理", en: "Auto Resource" },
-  auto_link: { zh: "自动关联", en: "Auto Link" },
-  auto_dream: { zh: "自动沉淀", en: "Auto Dream" },
-  proactive: { zh: "主动发现", en: "Proactive" },
-  reme_scene: { zh: "ReMe 应用场景", en: "ReMe Application Scenarios" },
-  framework: { zh: "ReMe 代码框架", en: "ReMe Framework" },
-  "reme-blog": { zh: "ReMe 博客", en: "ReMe Blog" },
-  contributing: { zh: "开源与贡献", en: "Open Source and Contributing" },
+const jobGroups = {
+  version: "system",
+  app_config: "system",
+  chat: "system",
+  health_check: "system",
+  status: "system",
+  help: "system",
+  auto_dream: "memory",
+  auto_memory: "memory",
+  auto_memory_cc: "memory",
+  auto_resource: "memory",
+  proactive: "memory",
+  traverse: "retrieval",
+  graph_snapshot: "retrieval",
+  reindex: "retrieval",
+  search: "retrieval",
+  node_search: "retrieval",
+  daily_list: "daily",
+  daily_reindex: "daily",
+  daily_write: "daily",
+  frontmatter_delete: "files",
+  frontmatter_read: "files",
+  frontmatter_update: "files",
+  stat: "files",
+  list: "files",
+  move: "files",
+  delete: "files",
+  read: "files",
+  load: "files",
+  read_image: "files",
+  write: "files",
+  save: "files",
+  edit: "files",
 };

-const productDocuments = [
-  {
-    slug: "typescript",
-    source: "typescript",
-    titles: { zh: "TypeScript Agent 集成", en: "TypeScript Agent Integrations" },
-    descriptions: {
-      zh: "配置统一 HTTP client，以及 DeepSeek Harness 和 OpenClaw 原生适配器。",
-      en: "Configure the shared HTTP client and native DeepSeek Harness and OpenClaw adapters.",
-    },
-    group: "integration",
-  },
-  {
-    slug: "studio",
-    source: "reme_studio",
-    titles: { zh: "ReMe 工作台", en: "ReMe Studio" },
-    descriptions: {
-      zh: "浏览、编辑和搜索本地记忆，并探索记忆图谱。",
-      en: "Browse, edit, search, and explore local memory from the web workspace.",
-    },
-    group: "workspace",
-  },
-  {
-    slug: "daily-paper",
-    source: "plugins/daily_paper",
-    titles: { zh: "每日论文插件", en: "Daily Paper Plugin" },
-    descriptions: {
-      zh: "发现论文、解析 PDF，并生成阅读笔记与每日简报。",
-      en: "Discover papers, analyze PDFs, and produce reading notes and a daily brief.",
-    },
-    group: "plugins",
-  },
-  {
-    slug: "auto-fin",
-    source: "plugins/auto-fin",
-    titles: { zh: "Auto Fin 插件", en: "Auto Fin Plugin" },
-    descriptions: {
-      zh: "结合最新财联社新闻与本地历史记忆生成研究报告。",
-      en: "Research recent CLS news with historical context from local memory.",
-    },
-    group: "plugins",
-  },
-  {
-    slug: "beam",
-    source: "benchmark/beam",
-    titles: { zh: "BEAM", en: "BEAM" },
-    descriptions: {
-      zh: "评测大规模记忆检索能力。",
-      en: "Evaluate memory retrieval at scale.",
-    },
-    group: "benchmarks",
-  },
-  {
-    slug: "longmemeval",
-    source: "benchmark/longmemeval",
-    titles: { zh: "LongMemEval", en: "LongMemEval" },
-    descriptions: {
-      zh: "评测跨会话长期记忆问答能力。",
-      en: "Evaluate long-term, cross-session memory question answering.",
-    },
-    group: "benchmarks",
-  },
-  {
-    slug: "pibench",
-    source: "benchmark/pibench",
-    titles: { zh: "π-Bench", en: "π-Bench" },
-    descriptions: {
-      zh: "评测带持久记忆的个人智能体。",
-      en: "Evaluate personal agents with persistent memory.",
-    },
-    group: "benchmarks",
-  },
-  {
-    slug: "toolmemory",
-    source: "benchmark/toolmemory",
-    titles: { zh: "Tool Memory / ExpG", en: "Tool Memory / ExpG" },
-    descriptions: {
-      zh: "通过经验驱动的自适应指导增强 Agent 工具使用。",
-      en: "Improve agent tool use through experience-driven adaptive guidance.",
-    },
-    group: "benchmarks",
-  },
-];
-
-const sharedDocuments = [
-  {
-    id: "agents-guide",
-    path: "AGENTS.md",
-    sourcePath: "AGENTS.md",
-    titles: {
-      zh: "Agent 开发指南",
-      en: "Agent Development Guide",
-    },
-    description: "Repository contracts, lifecycle rules, safety boundaries, and validation.",
-    group: "development",
-    language: "shared",
-  },
-];
-
-async function markdownTitle(filePath) {
-  const source = await readFile(filePath, "utf8");
-  return source.match(/^#\s+(.+)$/m)?.[1]?.replace(/[`*_]/g, "") || path.basename(filePath, ".md");
+function typeLabel(schema = {}) {
+  if (schema.oneOf) return schema.oneOf.map(typeLabel).join(" or ");
+  if (schema.type === "array") return `${typeLabel(schema.items || {})}[]`;
+  return schema.type || "any";
 }

-async function buildManifest() {
-  const documents = [
-    {
-      id: "readme-zh",
-      path: "README_ZH.md",
-      sourcePath: "README_ZH.md",
-      title: "ReMe 项目介绍",
-      description: "核心理念、快速开始、使用场景与社区入口。",
-      group: "overview",
-      language: "zh",
-    },
-    {
-      id: "readme-en",
-      path: "README.md",
-      sourcePath: "README.md",
-      title: "Introducing ReMe",
-      description: "Core ideas, quick start, use cases, and community resources.",
-      group: "overview",
-      language: "en",
-    },
-  ];
+function markdownCell(value) {
+  if (value === undefined) return "—";
+  const rendered = typeof value === "string" ? value : JSON.stringify(value);
+  return rendered
+    .replaceAll("<", "&lt;")
+    .replaceAll(">", "&gt;")
+    .replaceAll("|", "\\|")
+    .replaceAll("\n", " ");
+}

-  for (const language of ["zh", "en"]) {
-    for (const topic of topicOrder) {
-      const sourcePath = `docs/${language}/${topic}.md`;
-      documents.push({
-        id: `${language}-${topic}`,
-        path: sourcePath,
-        sourcePath,
-        title: localizedTitles[topic]?.[language] || (await markdownTitle(path.join(repoDir, sourcePath))),
-        description: "",
-        group: groups[topic],
-        language,
-      });
-    }
+function buildJobReference(config, language) {
+  const isZh = language === "zh";
+  const jobs = Object.entries(config.jobs || {}).filter(([, job]) => !["background", "cron"].includes(job.backend));
+  const sections = new Map();

-    for (const product of productDocuments) {
-      const filename = language === "zh" ? "README_ZH.md" : "README.md";
-      documents.push({
-        id: `${product.slug}-${language}`,
-        path: `${product.source}/${filename}`,
-        sourcePath: `${product.source}/${filename}`,
-        title: product.titles[language],
-        description: product.descriptions[language],
-        group: product.group,
-        language,
-      });
-    }
+  for (const [name, job] of jobs) {
+    const group = jobGroups[name] || "system";
+    if (!sections.has(group)) sections.set(group, []);
+    sections.get(group).push([name, job]);
  }

-  return [...documents, ...sharedDocuments].sort(
-    (left, right) => navigationGroupOrder.indexOf(left.group) - navigationGroupOrder.indexOf(right.group),
-  );
+  const lines = [
+    "---",
+    `title: ${isZh ? "Job API 参考" : "Job API Reference"}`,
+    `description: ${isZh ? "从默认配置自动生成的可调用 Job、参数和服务边界。" : "Callable jobs, parameters, and service boundaries generated from the default configuration."}`,
+    "---",
+    "",
+    `# ${isZh ? "Job API 参考" : "Job API Reference"}`,
+    "",
+    isZh
+      ? "本页从 `reme/config/default.yaml` 自动生成。它描述默认应用中的可调用 Job；插件和自定义配置可以增加、删除或覆盖 Job。运行 `reme help` 可查看当前服务的实际能力。"
+      : "This page is generated from `reme/config/default.yaml`. It describes callable jobs in the default application; plugins and custom configurations may add, remove, or override jobs. Run `reme help` to inspect the active service.",
+    "",
+    isZh
+      ? "> 后台 Job 和 Cron Job 不通过服务暴露，因此不列入调用参考。"
+      : "> Background and cron jobs are not service-exposed and are omitted from the callable reference.",
+    "",
+  ];
+
+  for (const [group, entries] of sections) {
+    lines.push(`## ${groupNames[language][group]}`, "");
+    for (const [name, job] of entries) {
+      const properties = job.parameters?.properties || {};
+      const required = new Set(job.parameters?.required || []);
+      lines.push(`### \`${name}\``, "", markdownCell(job.description || ""), "");
+      lines.push("```bash", `reme ${name}${Object.keys(properties).length ? " ..." : ""}`, "```", "");
+      if (!Object.keys(properties).length) {
+        lines.push(isZh ? "无参数。" : "No parameters.", "");
+        continue;
+      }
+      lines.push(
+        isZh
+          ? "| 参数 | 类型 | 必填 | 默认值 | 说明 |"
+          : "| Parameter | Type | Required | Default | Description |",
+        "|---|---|---:|---|---|",
+      );
+      for (const [parameter, schema] of Object.entries(properties)) {
+        lines.push(
+          `| \`${parameter}\` | \`${markdownCell(typeLabel(schema))}\` | ${required.has(parameter) ? (isZh ? "是" : "yes") : (isZh ? "否" : "no")} | ${markdownCell(schema.default)} | ${markdownCell(schema.description || "—")} |`,
+        );
+      }
+      lines.push("");
+    }
+  }
+  return `${lines.join("\n")}\n`;
 }

 await rm(path.join(siteDir, ".generated"), { recursive: true, force: true });
 await mkdir(outputDir, { recursive: true });
-await cp(path.join(siteDir, "public", "favicon.svg"), path.join(siteDir, ".generated", "favicon.svg"));
-await cp(path.join(siteDir, "public", "CNAME"), path.join(siteDir, ".generated", "CNAME"));
-
-for (const file of ["README.md", "README_ZH.md", "AGENTS.md"]) {
-  await cp(path.join(repoDir, file), path.join(outputDir, file));
-}
-await cp(path.join(repoDir, "docs"), path.join(outputDir, "docs"), {
+await cp(path.join(repoDir, "docs"), outputDir, {
  recursive: true,
-  filter: (source) => path.basename(source) !== ".DS_Store",
+  filter: (source) => ![".DS_Store", "plans"].includes(path.basename(source)),
 });
-for (const product of productDocuments) {
-  await mkdir(path.join(outputDir, product.source), { recursive: true });
-  for (const filename of ["README.md", "README_ZH.md"]) {
-    await cp(path.join(repoDir, product.source, filename), path.join(outputDir, product.source, filename));
-  }
+await cp(path.join(siteDir, "public", "CNAME"), path.join(outputDir, "public", "CNAME"));
+
+const sourceMap = {};
+for (const [destination, source] of externalDocuments) {
+  const destinationPath = path.join(outputDir, destination);
+  await mkdir(path.dirname(destinationPath), { recursive: true });
+  await cp(path.join(repoDir, source), destinationPath);
+  sourceMap[destination] = source;
 }
-await writeFile(
-  path.join(outputDir, "manifest.json"),
-  `${JSON.stringify({ documents: await buildManifest() }, null, 2)}\n`,
-);
+
+for (const language of ["zh", "en"]) {
+  await cp(
+    path.join(repoDir, "benchmark/toolmemory/gitcha.png"),
+    path.join(outputDir, language, "benchmarks", "gitcha.png"),
+  );
+}
+
+const defaultConfig = parseYaml(await readFile(path.join(repoDir, "reme/config/default.yaml"), "utf8"));
+for (const language of ["zh", "en"]) {
+  const destination = path.join(outputDir, language, "reference", "jobs.md");
+  await mkdir(path.dirname(destination), { recursive: true });
+  await writeFile(destination, buildJobReference(defaultConfig, language));
+  sourceMap[`${language}/reference/jobs.md`] = "reme/config/default.yaml";
+}
+
+await writeFile(path.join(outputDir, ".source-map.json"), `${JSON.stringify(sourceMap, null, 2)}\n`);
--- a/github-pages/scripts/verify-build.mjs
+++ b/github-pages/scripts/verify-build.mjs
@ -0,0 +1,91 @@
+import assert from "node:assert/strict";
+import { access, readFile, readdir } from "node:fs/promises";
+import path from "node:path";
+import { fileURLToPath } from "node:url";
+
+const siteDir = path.resolve(path.dirname(fileURLToPath(import.meta.url)), "..");
+const outputDir = path.join(siteDir, "dist");
+
+async function collectFiles(directory, prefix = "") {
+  const files = [];
+  for (const entry of await readdir(directory, { withFileTypes: true })) {
+    const relativePath = path.posix.join(prefix, entry.name);
+    if (entry.isDirectory()) files.push(...await collectFiles(path.join(directory, entry.name), relativePath));
+    else files.push(relativePath);
+  }
+  return files;
+}
+
+function pageUrl(relativePath) {
+  if (relativePath === "index.html") return "/";
+  if (relativePath.endsWith("/index.html")) return `/${relativePath.slice(0, -"index.html".length)}`;
+  return `/${relativePath.slice(0, -".html".length)}`;
+}
+
+function routeExists(pathname, files) {
+  const relativePath = decodeURIComponent(pathname).replace(/^\/+/, "");
+  if (!relativePath) return files.has("index.html");
+  if (relativePath.endsWith("/")) return files.has(`${relativePath}index.html`);
+  return files.has(relativePath) || files.has(`${relativePath}.html`) || files.has(`${relativePath}/index.html`);
+}
+
+const requiredFiles = [
+  "index.html",
+  "404.html",
+  "CNAME",
+  "favicon.svg",
+  "hashmap.json",
+  "sitemap.xml",
+  "llms.txt",
+  "llms-full.txt",
+  "zh/index.html",
+  "en/index.html",
+  "zh/configuration.html",
+  "en/configuration.html",
+  "zh/services.html",
+  "en/services.html",
+  "zh/reference/jobs.html",
+  "en/reference/jobs.html",
+  "zh/configuration/llms.txt",
+  "en/configuration/llms.txt",
+];
+
+for (const relativePath of requiredFiles) await access(path.join(outputDir, relativePath));
+
+assert.equal((await readFile(path.join(outputDir, "CNAME"), "utf8")).trim(), "reme.agentscope.io");
+
+const homepage = await readFile(path.join(outputDir, "index.html"), "utf8");
+assert.ok(homepage.includes('href="/en/"'), "root language switch must link to /en/");
+assert.ok(!homepage.includes('href="/en/ex"'), "root language switch must not produce /en/ex");
+assert.ok(homepage.includes('"studio-en":"/en/workspace/studio"'), "legacy redirects must be embedded");
+
+const sitemap = await readFile(path.join(outputDir, "sitemap.xml"), "utf8");
+assert.ok(sitemap.includes("<lastmod>"), "sitemap must include canonical-source update times");
+assert.ok(!sitemap.includes("<loc>https://reme.agentscope.io/</loc>"), "root redirect must not be indexed");
+assert.ok(!sitemap.includes('hreflang="zh-CN" href="https://reme.agentscope.io/"'), "root must not duplicate zh-CN");
+
+const ChineseConfiguration = await readFile(path.join(outputDir, "zh/configuration.html"), "utf8");
+assert.match(ChineseConfiguration, /搜索文档/);
+assert.match(ChineseConfiguration, /复制 Markdown/);
+assert.match(ChineseConfiguration, /在 GitHub 查看源文件/);
+
+const jobReference = await readFile(path.join(outputDir, "en/reference/jobs.html"), "utf8");
+assert.match(jobReference, /Job API Reference/);
+assert.match(jobReference, /auto_memory/);
+
+const outputFiles = new Set(await collectFiles(outputDir));
+const missingLinks = [];
+for (const relativePath of [...outputFiles].filter((file) => file.endsWith(".html"))) {
+  const html = await readFile(path.join(outputDir, relativePath), "utf8");
+  const currentUrl = new URL(pageUrl(relativePath), "https://reme-docs.local");
+  for (const match of html.matchAll(/<a\b[^>]*\bhref="([^"]+)"/g)) {
+    const href = match[1].replaceAll("&amp;", "&");
+    if (href.startsWith("#")) continue;
+    const target = new URL(href, currentUrl);
+    if (target.origin !== currentUrl.origin) continue;
+    if (!routeExists(target.pathname, outputFiles)) missingLinks.push(`${relativePath}: ${href}`);
+  }
+}
+assert.deepEqual(missingLinks, [], `missing internal links:\n${missingLinks.join("\n")}`);
+
+console.log(`Verified ${requiredFiles.length} documentation build artifacts.`);
--- a/github-pages/src/main.js
+++ b/github-pages/src/main.js
@ -1,458 +0,0 @@
-import DOMPurify from "dompurify";
-import { marked } from "marked";
-import { stripMarkdownFrontmatter } from "./markdown.js";
-import "./styles.css";
-
-const baseUrl = import.meta.env.BASE_URL;
-const repositoryUrl = "https://github.com/agentscope-ai/ReMe";
-
-const copy = {
-  zh: {
-    docs: "文档",
-    home: "首页",
-    search: "搜索文档…",
-    noResults: "没有找到匹配的文档",
-    menu: "打开导航",
-    toc: "本页目录",
-    edit: "在 GitHub 查看源文件",
-    quickStart: "快速开始",
-    groups: {
-      overview: "项目介绍",
-      start: "开始使用",
-      fundamentals: "核心原理",
-      automation: "自动化能力",
-      concepts: "架构与场景",
-      integration: "Agent 集成",
-      workspace: "工作区",
-      plugins: "插件",
-      benchmarks: "评测",
-      development: "开发者规范",
-    },
-  },
-  en: {
-    docs: "Documentation",
-    home: "Home",
-    search: "Search documentation…",
-    noResults: "No matching documents",
-    menu: "Open navigation",
-    toc: "On this page",
-    edit: "View source on GitHub",
-    quickStart: "Quick start",
-    groups: {
-      overview: "Introduction",
-      start: "Get started",
-      fundamentals: "Fundamentals",
-      automation: "Automation",
-      concepts: "Architecture & scenarios",
-      integration: "Agent integration",
-      workspace: "Workspace",
-      plugins: "Plugins",
-      benchmarks: "Benchmarks",
-      development: "Development",
-    },
-  },
-};
-
-const state = {
-  language: localStorage.getItem("reme-docs-language") || "zh",
-  documents: [],
-  activeDocument: null,
-  query: "",
-};
-
-const homeCopy = {
-  zh: {
-    eyebrow: "LOCAL-FIRST · FILE-NATIVE",
-    title: "让 Agent 真正记住，\n也让记忆始终属于你。",
-    description: "ReMe 将对话和资料沉淀为可读、可编辑、可检索、相互链接的 Markdown，并提供从工作区管理到主动研究的一整套工具。",
-    start: "快速开始",
-    project: "了解 ReMe",
-    explore: "按目标探索",
-    exploreDescription: "选择你现在想完成的事情。每个入口都直接连接到对应的完整文档。",
-    cards: [
-      { id: "studio-zh", icon: "◫", label: "管理记忆", title: "ReMe 工作台", description: "在本地 Web 工作区中浏览、编辑、搜索记忆，并探索 wikilink 图谱。", tone: "mint" },
-      { id: "daily-paper-zh", icon: "◌", label: "发现与分析", title: "每日论文插件", description: "从论文榜单筛选值得阅读的工作，解析 PDF，并生成笔记与五分钟简报。", tone: "blue" },
-      { id: "auto-fin-zh", icon: "↗", label: "主题研究", title: "Auto Fin 插件", description: "连接最新财联社新闻和本地历史记忆，生成可追溯的研究报告。", tone: "amber" },
-    ],
-    benchmark: "验证记忆能力",
-    benchmarkDescription: "从检索规模、跨会话问答、个人智能体到工具经验，查看 ReMe 的四套评测与复现实验。",
-    benchmarkAction: "从 BEAM 开始",
-  },
-  en: {
-    eyebrow: "LOCAL-FIRST · FILE-NATIVE",
-    title: "Memory that works for agents.\nFiles that remain yours.",
-    description: "ReMe turns conversations and resources into readable, editable, searchable, interconnected Markdown—with tools spanning workspace management and proactive research.",
-    start: "Quick start",
-    project: "Meet ReMe",
-    explore: "Explore by goal",
-    exploreDescription: "Start with what you want to accomplish. Every entry opens the complete guide.",
-    cards: [
-      { id: "studio-en", icon: "◫", label: "Manage memory", title: "ReMe Studio", description: "Browse, edit, and search memory in a local web workspace, then explore its wikilink graph.", tone: "mint" },
-      { id: "daily-paper-en", icon: "◌", label: "Discover & analyze", title: "Daily Paper Plugin", description: "Select useful papers from rankings, analyze PDFs, and create notes plus a five-minute brief.", tone: "blue" },
-      { id: "auto-fin-en", icon: "↗", label: "Research topics", title: "Auto Fin Plugin", description: "Connect recent CLS news with historical local memory to produce traceable research reports.", tone: "amber" },
-    ],
-    benchmark: "Validate memory systems",
-    benchmarkDescription: "Explore four reproducible evaluations covering retrieval scale, cross-session QA, personal agents, and tool-use experience.",
-    benchmarkAction: "Start with BEAM",
-  },
-};
-
-const app = document.querySelector("#app");
-
-app.innerHTML = `
-  <header class="topbar">
-    <a class="brand" href="${baseUrl}" aria-label="ReMe documentation home">
-      <span class="brand-mark">R</span>
-      <span>ReMe</span>
-      <span class="brand-divider"></span>
-      <span class="brand-section" data-copy="docs"></span>
-    </a>
-    <nav class="top-actions" aria-label="Global navigation">
-      <div class="language-switch" role="group" aria-label="Language">
-        <button type="button" data-language="zh">中</button>
-        <button type="button" data-language="en">EN</button>
-      </div>
-      <a class="quick-start-link" href="?doc=zh-quick_start" data-doc="zh-quick_start" data-copy="quickStart"></a>
-      <a class="github-link" href="${repositoryUrl}" target="_blank" rel="noreferrer">GitHub ↗</a>
-      <button class="menu-button" type="button" aria-expanded="false" data-action="menu"></button>
-    </nav>
-  </header>
-  <div class="docs-shell">
-    <aside class="sidebar" aria-label="Documentation navigation">
-      <label class="search-box">
-        <span aria-hidden="true">⌕</span>
-        <input type="search" autocomplete="off" />
-        <kbd>⌘K</kbd>
-      </label>
-      <nav class="document-nav"></nav>
-      <div class="sidebar-footer">
-        <span class="status-dot"></span>
-        Local-first · File-native
-      </div>
-    </aside>
-    <main class="article-wrap">
-      <article class="article"><div class="loading-line"></div></article>
-    </main>
-    <aside class="toc-panel"><nav class="toc"></nav></aside>
-  </div>
-  <button class="sidebar-backdrop" type="button" aria-label="Close navigation"></button>
-`;
-
-const sidebar = app.querySelector(".sidebar");
-const docsShell = app.querySelector(".docs-shell");
-const documentNav = app.querySelector(".document-nav");
-const article = app.querySelector(".article");
-const toc = app.querySelector(".toc");
-const searchInput = app.querySelector("input[type='search']");
-const menuButton = app.querySelector(".menu-button");
-const backdrop = app.querySelector(".sidebar-backdrop");
-
-function slugify(value) {
-  return value
-    .toLowerCase()
-    .trim()
-    .replace(/<[^>]+>/g, "")
-    .replace(/[^\p{Letter}\p{Number}]+/gu, "-")
-    .replace(/^-|-$/g, "");
-}
-
-function resolveDocumentPath(currentPath, target) {
-  const cleanTarget = target.split("#")[0].split("?")[0];
-  const currentParts = currentPath.split("/");
-  currentParts.pop();
-  for (const part of cleanTarget.split("/")) {
-    if (!part || part === ".") continue;
-    if (part === "..") currentParts.pop();
-    else currentParts.push(part);
-  }
-  return currentParts.join("/");
-}
-
-function configureMarkdown(document) {
-  const renderer = new marked.Renderer();
-  const headingIds = new Map();
-
-  renderer.heading = ({ tokens, depth }) => {
-    const text = tokens.map((token) => token.text || token.raw || "").join("");
-    const baseSlug = slugify(text) || "section";
-    const count = headingIds.get(baseSlug) || 0;
-    headingIds.set(baseSlug, count + 1);
-    const id = count ? `${baseSlug}-${count + 1}` : baseSlug;
-    return `<h${depth} id="${id}">${text}</h${depth}>`;
-  };
-
-  renderer.image = ({ href, title, text }) => {
-    const url = /^(https?:|data:)/.test(href)
-      ? href
-      : `${baseUrl}content/${resolveDocumentPath(document.path, href)}`;
-    const titleAttribute = title ? ` title="${title}"` : "";
-    return `<img src="${url}" alt="${text}" loading="lazy"${titleAttribute}>`;
-  };
-
-  renderer.link = ({ href, title, tokens }) => {
-    const label = tokens.map((token) => token.text || token.raw || "").join("");
-    const titleAttribute = title ? ` title="${title}"` : "";
-    if (href.startsWith("#")) return `<a href="${href}"${titleAttribute}>${label}</a>`;
-    if (!/^(https?:|mailto:)/.test(href)) {
-      const resolved = resolveDocumentPath(document.path, href);
-      const localDocument = state.documents.find((item) => item.path === resolved);
-      if (localDocument) return `<a href="?doc=${localDocument.id}" data-doc="${localDocument.id}">${label}</a>`;
-      return `<a href="${repositoryUrl}/blob/main/${resolved}" target="_blank" rel="noreferrer">${label}</a>`;
-    }
-    return `<a href="${href}" target="_blank" rel="noreferrer"${titleAttribute}>${label}</a>`;
-  };
-
-  marked.use({ renderer, gfm: true, breaks: false });
-}
-
-function availableDocuments() {
-  return state.documents.filter(
-    (document) => document.language === state.language || document.language === "shared",
-  );
-}
-
-function documentTitle(document) {
-  return document.titles?.[state.language] || document.title;
-}
-
-function renderChrome() {
-  const labels = copy[state.language];
-  app.querySelector("[data-copy='docs']").textContent = labels.docs;
-  const quickStartLink = app.querySelector("[data-copy='quickStart']");
-  quickStartLink.textContent = `${labels.quickStart} →`;
-  quickStartLink.href = `?doc=${state.language}-quick_start`;
-  quickStartLink.dataset.doc = `${state.language}-quick_start`;
-  searchInput.placeholder = labels.search;
-  menuButton.textContent = labels.menu;
-  document.documentElement.lang = state.language === "zh" ? "zh-CN" : "en";
-  app.querySelectorAll("[data-language]").forEach((button) => {
-    button.classList.toggle("active", button.dataset.language === state.language);
-  });
-}
-
-function renderNavigation() {
-  const labels = copy[state.language];
-  const query = state.query.trim().toLocaleLowerCase();
-  const filtered = availableDocuments().filter((document) =>
-    `${documentTitle(document)} ${document.title || ""} ${document.description}`.toLocaleLowerCase().includes(query),
-  );
-  const groups = [...new Set(filtered.map((document) => document.group))];
-
-  if (!filtered.length) {
-    documentNav.innerHTML = `
-      <a href="${baseUrl}" data-home class="home-link ${state.activeDocument ? "" : "active"}">${labels.home}</a>
-      <p class="empty-state">${labels.noResults}</p>
-    `;
-    return;
-  }
-
-  documentNav.innerHTML = `
-    <a href="${baseUrl}" data-home class="home-link ${state.activeDocument ? "" : "active"}">${labels.home}</a>
-  ` + groups
-    .map(
-      (group) => `
-        <section class="nav-group">
-          <h2>${labels.groups[group]}</h2>
-          ${filtered
-            .filter((document) => document.group === group)
-            .map(
-              (document) => `
-                <a href="?doc=${document.id}" data-doc="${document.id}" class="${state.activeDocument?.id === document.id ? "active" : ""}">
-                  <span>${documentTitle(document)}</span>
-                </a>`,
-            )
-            .join("")}
-        </section>`,
-    )
-    .join("");
-}
-
-function renderHome(pushHistory = true) {
-  const labels = homeCopy[state.language];
-  state.activeDocument = null;
-  docsShell.classList.add("home-view");
-  article.dataset.group = "home";
-  renderNavigation();
-  article.innerHTML = `
-    <section class="home-hero">
-      <p class="home-eyebrow">${labels.eyebrow}</p>
-      <h1>${labels.title.replace("\n", "<br>")}</h1>
-      <p class="home-lead">${labels.description}</p>
-      <div class="home-actions">
-        <a href="${baseUrl}?doc=${state.language}-quick_start" class="primary-action">${labels.start} →</a>
-        <a href="${baseUrl}?doc=readme-${state.language}" class="secondary-action">${labels.project}</a>
-      </div>
-    </section>
-    <section class="home-explore">
-      <p class="section-kicker">01 / PRODUCT & PLUGINS</p>
-      <h2>${labels.explore}</h2>
-      <p class="section-lead">${labels.exploreDescription}</p>
-      <div class="feature-grid">
-        ${labels.cards.map((card) => `
-          <a href="${baseUrl}?doc=${card.id}" class="feature-card ${card.tone}">
-            <span class="feature-icon">${card.icon}</span>
-            <span class="feature-label">${card.label}</span>
-            <strong>${card.title}</strong>
-            <span class="feature-description">${card.description}</span>
-            <span class="feature-arrow">→</span>
-          </a>`).join("")}
-      </div>
-    </section>
-    <section class="benchmark-callout">
-      <div>
-        <p class="section-kicker">02 / BENCHMARKS</p>
-        <h2>${labels.benchmark}</h2>
-        <p>${labels.benchmarkDescription}</p>
-      </div>
-      <a href="${baseUrl}?doc=beam-${state.language}">${labels.benchmarkAction} →</a>
-    </section>
-  `;
-  toc.innerHTML = "";
-  closeMenu();
-  if (pushHistory) history.pushState({ home: true }, "", baseUrl);
-  window.scrollTo({ top: 0, behavior: "instant" });
-}
-
-function renderToc() {
-  const headings = [...article.querySelectorAll("h2, h3")];
-  if (!headings.length) {
-    toc.innerHTML = "";
-    return;
-  }
-  toc.innerHTML = `
-    <h2>${copy[state.language].toc}</h2>
-    ${headings
-      .map(
-        (heading) => `<a class="toc-${heading.tagName.toLowerCase()}" href="#${heading.id}">${heading.childNodes[0]?.textContent || heading.textContent}</a>`,
-      )
-      .join("")}
-  `;
-}
-
-function rewriteRenderedUrls(document) {
-  article.querySelectorAll("img[src]").forEach((image) => {
-    const source = image.getAttribute("src");
-    if (source && !/^(https?:|data:|\/)/.test(source)) {
-      image.src = `${baseUrl}content/${resolveDocumentPath(document.path, source)}`;
-    }
-  });
-
-  article.querySelectorAll("a[href]").forEach((link) => {
-    const href = link.getAttribute("href");
-    if (!href || /^(https?:|mailto:|#|\/)/.test(href) || link.dataset.doc) return;
-    const resolved = resolveDocumentPath(document.path, href);
-    const localDocument = state.documents.find((item) => item.path === resolved);
-    if (localDocument) {
-      link.href = `?doc=${localDocument.id}`;
-      link.dataset.doc = localDocument.id;
-      link.removeAttribute("target");
-      return;
-    }
-    link.href = `${repositoryUrl}/blob/main/${resolved}`;
-    link.target = "_blank";
-    link.rel = "noreferrer";
-  });
-}
-
-async function openDocument(id, pushHistory = true) {
-  const fallbackId = state.language === "zh" ? "readme-zh" : "readme-en";
-  const document = state.documents.find((item) => item.id === id) || state.documents.find((item) => item.id === fallbackId);
-  if (document.language !== "shared" && document.language !== state.language) {
-    state.language = document.language;
-    localStorage.setItem("reme-docs-language", state.language);
-    renderChrome();
-  }
-  state.activeDocument = document;
-  docsShell.classList.remove("home-view");
-  article.dataset.group = document.group;
-  renderNavigation();
-  article.innerHTML = `<div class="loading-line"></div>`;
-
-  const response = await fetch(`${baseUrl}content/${document.path}`);
-  if (!response.ok) throw new Error(`Unable to load ${document.path}`);
-  configureMarkdown(document);
-  const markdown = stripMarkdownFrontmatter(await response.text());
-  const body = DOMPurify.sanitize(await marked.parse(markdown), {
-    ADD_ATTR: ["target"],
-  });
-  article.innerHTML = `
-    <div class="article-meta">
-      <span>${copy[state.language].groups[document.group]}</span>
-      <span>·</span>
-      <span>${document.sourcePath}</span>
-    </div>
-    <div class="markdown-body">${body}</div>
-    <footer class="article-footer">
-      <a href="${repositoryUrl}/blob/main/${document.sourcePath}" target="_blank" rel="noreferrer">${copy[state.language].edit} ↗</a>
-    </footer>
-  `;
-  rewriteRenderedUrls(document);
-  renderToc();
-  closeMenu();
-  if (pushHistory) history.pushState({ doc: document.id }, "", `?doc=${document.id}`);
-  window.scrollTo({ top: 0, behavior: "instant" });
-}
-
-function closeMenu() {
-  sidebar.classList.remove("open");
-  backdrop.classList.remove("visible");
-  menuButton.setAttribute("aria-expanded", "false");
-}
-
-function toggleMenu() {
-  const open = !sidebar.classList.contains("open");
-  sidebar.classList.toggle("open", open);
-  backdrop.classList.toggle("visible", open);
-  menuButton.setAttribute("aria-expanded", String(open));
-}
-
-app.addEventListener("click", (event) => {
-  const homeLink = event.target.closest("[data-home]");
-  if (homeLink) {
-    event.preventDefault();
-    renderHome();
-    return;
-  }
-  const documentLink = event.target.closest("[data-doc]");
-  if (documentLink) {
-    event.preventDefault();
-    openDocument(documentLink.dataset.doc);
-  }
-  if (event.target.closest("[data-action='menu']")) toggleMenu();
-  const languageButton = event.target.closest("[data-language]");
-  if (languageButton && languageButton.dataset.language !== state.language) {
-    state.language = languageButton.dataset.language;
-    localStorage.setItem("reme-docs-language", state.language);
-    state.query = "";
-    searchInput.value = "";
-    renderChrome();
-    renderHome();
-  }
-});
-
-searchInput.addEventListener("input", () => {
-  state.query = searchInput.value;
-  renderNavigation();
-});
-
-document.addEventListener("keydown", (event) => {
-  if ((event.metaKey || event.ctrlKey) && event.key.toLowerCase() === "k") {
-    event.preventDefault();
-    searchInput.focus();
-  }
-  if (event.key === "Escape") closeMenu();
-});
-
-backdrop.addEventListener("click", closeMenu);
-window.addEventListener("popstate", (event) => {
-  const id = event.state?.doc || new URLSearchParams(location.search).get("doc");
-  if (id) openDocument(id, false);
-  else renderHome(false);
-});
-
-const manifest = await fetch(`${baseUrl}content/manifest.json`).then((response) => response.json());
-state.documents = manifest.documents;
-renderChrome();
-const initialDocument = new URLSearchParams(location.search).get("doc");
-if (initialDocument) await openDocument(initialDocument, false);
-else renderHome(false);
--- a/github-pages/src/markdown.js
+++ b/github-pages/src/markdown.js
@ -1,6 +0,0 @@
-const FRONTMATTER_PATTERN = /^\uFEFF?---\r?\n[\s\S]*?\r?\n---(?:\r?\n|$)/;
-
-/** Remove a leading YAML frontmatter block before rendering Markdown. */
-export function stripMarkdownFrontmatter(markdown) {
-  return markdown.replace(FRONTMATTER_PATTERN, "");
-}
--- a/github-pages/src/styles.css
+++ b/github-pages/src/styles.css
@ -1,200 +0,0 @@
-:root {
-  color: #17221d;
-  background: #f4f7f5;
-  font-family: Inter, ui-sans-serif, system-ui, -apple-system, BlinkMacSystemFont, "Segoe UI", sans-serif;
-  font-synthesis: none;
-  text-rendering: optimizeLegibility;
-  --ink: #17221d;
-  --muted: #65716a;
-  --line: #dce5e0;
-  --paper: #ffffff;
-  --green: #087f6a;
-  --blue: #3156d9;
-  --green-soft: #e5f5ef;
-  --code: #f3f5f2;
-}
-
-* { box-sizing: border-box; }
-html { scroll-behavior: smooth; scroll-padding-top: 92px; }
-body {
-  margin: 0;
-  min-width: 320px;
-  background:
-    radial-gradient(circle at 8% 9%, rgba(27, 193, 164, 0.06), transparent 26rem),
-    #f4f7f5;
-}
-button, input { font: inherit; }
-a { color: inherit; }
-
-.topbar {
-  position: fixed;
-  inset: 0 0 auto 0;
-  z-index: 30;
-  height: 66px;
-  display: flex;
-  align-items: center;
-  justify-content: space-between;
-  padding: 0 28px;
-  border-bottom: 1px solid rgba(213, 225, 219, 0.85);
-  background: rgba(250, 252, 251, 0.88);
-  backdrop-filter: blur(18px) saturate(140%);
-}
-
-.brand { display: flex; align-items: center; gap: 11px; color: var(--ink); text-decoration: none; font-weight: 760; }
-.brand-mark {
-  display: grid;
-  place-items: center;
-  width: 34px;
-  height: 34px;
-  border-radius: 11px;
-  color: white;
-  background: linear-gradient(145deg, #19c9b0, #3156d9 82%);
-  box-shadow: 0 7px 18px rgba(24, 123, 114, 0.2);
-  font-family: Georgia, serif;
-  font-size: 21px;
-}
-.brand-divider { width: 1px; height: 20px; background: var(--line); margin-left: 2px; }
-.brand-section { color: var(--muted); font-weight: 520; }
-.top-actions { display: flex; align-items: center; gap: 18px; }
-.github-link, .quick-start-link { color: #37443d; text-decoration: none; font-size: 13px; font-weight: 650; }
-.github-link:hover, .quick-start-link:hover { color: var(--green); }
-.quick-start-link { padding: 8px 12px; border: 1px solid #d7e5df; border-radius: 9px; background: rgba(255, 255, 255, 0.72); }
-.language-switch { display: flex; padding: 3px; border: 1px solid var(--line); border-radius: 9px; background: #f5f7f4; }
-.language-switch button { padding: 5px 9px; border: 0; border-radius: 6px; color: var(--muted); background: transparent; cursor: pointer; font-size: 12px; font-weight: 700; }
-.language-switch button.active { color: var(--ink); background: white; box-shadow: 0 1px 3px rgba(20, 40, 30, 0.1); }
-.menu-button { display: none; border: 1px solid var(--line); border-radius: 8px; background: white; padding: 7px 10px; cursor: pointer; }
-
-.docs-shell { display: grid; grid-template-columns: 276px minmax(0, 1fr) 224px; max-width: 1540px; min-height: 100vh; margin: 0 auto; padding-top: 66px; }
-.docs-shell.home-view { grid-template-columns: 276px minmax(0, 1fr); }
-.docs-shell.home-view .toc-panel { display: none; }
-.sidebar { position: sticky; top: 66px; height: calc(100vh - 66px); padding: 25px 20px 18px; overflow-y: auto; border-right: 1px solid var(--line); background: rgba(247, 250, 248, 0.78); }
-.search-box { display: flex; align-items: center; gap: 8px; height: 41px; padding: 0 11px; border: 1px solid #d8e3dd; border-radius: 11px; color: #7a857e; background: rgba(255, 255, 255, 0.84); box-shadow: 0 5px 18px rgba(29, 65, 48, 0.035); }
-.search-box:focus-within { border-color: #78aa8e; box-shadow: 0 0 0 3px rgba(22, 120, 76, 0.1); }
-.search-box input { width: 100%; border: 0; outline: 0; color: var(--ink); background: transparent; font-size: 13px; }
-.search-box kbd { padding: 2px 5px; border: 1px solid var(--line); border-radius: 4px; background: #f7f8f6; font-size: 10px; }
-.document-nav { padding: 12px 0 52px; }
-.home-link { position: relative; display: block; margin-top: 9px; padding: 9px 10px 9px 13px; border-radius: 8px; color: #536058; text-decoration: none; font-size: 13px; }
-.home-link:hover { color: var(--ink); background: #f0f3ef; }
-.home-link.active { color: #086b5a; background: linear-gradient(90deg, #dff3ec, #eaf6f2); font-weight: 700; }
-.nav-group { margin-top: 21px; }
-.nav-group h2 { margin: 0 10px 7px; color: #8a948e; font-size: 10px; font-weight: 800; letter-spacing: 0.11em; text-transform: uppercase; }
-.nav-group a { position: relative; display: block; padding: 7px 10px 7px 13px; border-radius: 8px; color: #536058; text-decoration: none; font-size: 13px; line-height: 1.4; }
-.nav-group a:hover { color: var(--ink); background: #f0f3ef; }
-.nav-group a.active { color: #086b5a; background: linear-gradient(90deg, #dff3ec, #eaf6f2); font-weight: 700; }
-.nav-group a.active::before { position: absolute; top: 9px; bottom: 9px; left: 0; width: 3px; border-radius: 3px; background: linear-gradient(#17b79d, #3470d8); content: ""; }
-.empty-state { padding: 24px 10px; color: var(--muted); font-size: 13px; }
-.sidebar-footer { position: sticky; bottom: -18px; display: flex; align-items: center; gap: 8px; margin: 0 -20px; padding: 14px 22px 18px; border-top: 1px solid var(--line); color: #7c8880; background: #f7faf8; font: 600 10px/1.2 ui-monospace, SFMono-Regular, Menlo, monospace; letter-spacing: 0.04em; text-transform: uppercase; }
-.status-dot { width: 6px; height: 6px; border-radius: 50%; background: #29a869; box-shadow: 0 0 0 3px #dff3e8; }
-
-.article-wrap { min-width: 0; padding: 58px clamp(32px, 5.8vw, 86px) 100px; background: rgba(255, 255, 255, 0.94); }
-.article { width: 100%; max-width: 820px; margin: 0 auto; }
-.article[data-group="home"] { max-width: 980px; }
-.home-hero { padding: 34px 0 76px; }
-.home-eyebrow, .section-kicker { margin: 0 0 17px; color: #12806d; font: 750 11px/1.4 ui-monospace, SFMono-Regular, Menlo, monospace; letter-spacing: 0.13em; }
-.home-hero h1 { margin: 0; color: #102019; font-size: clamp(46px, 6.3vw, 76px); line-height: 1.15; letter-spacing: -0.055em; }
-.home-lead { margin: 27px 0 0; color: #526159; font-size: 18px; line-height: 1.72; }
-.home-actions { display: flex; flex-wrap: wrap; gap: 11px; margin-top: 31px; }
-.home-actions a { padding: 11px 17px; border-radius: 10px; text-decoration: none; font-size: 14px; font-weight: 720; }
-.primary-action { color: white; background: #087f6a; box-shadow: 0 8px 22px rgba(8, 127, 106, 0.2); }
-.secondary-action { border: 1px solid #d5e2dc; color: #34443c; background: white; }
-.home-explore { padding-top: 58px; border-top: 1px solid var(--line); }
-.home-explore h2, .benchmark-callout h2 { margin: 0; color: #15251d; font-size: 30px; letter-spacing: -0.025em; }
-.section-lead { max-width: 620px; margin: 10px 0 25px; color: var(--muted); line-height: 1.65; }
-.feature-grid { display: grid; grid-template-columns: repeat(3, 1fr); gap: 14px; }
-.feature-card { position: relative; display: flex; min-height: 260px; flex-direction: column; padding: 23px; overflow: hidden; border: 1px solid #dce8e2; border-radius: 16px; color: var(--ink); background: linear-gradient(155deg, #fff, #f4faf7); text-decoration: none; transition: transform 160ms ease, box-shadow 160ms ease; }
-.feature-card:hover { transform: translateY(-3px); box-shadow: 0 16px 34px rgba(29, 68, 49, 0.1); }
-.feature-card.blue { background: linear-gradient(155deg, #fff, #f1f5ff); }
-.feature-card.amber { background: linear-gradient(155deg, #fff, #fbf7ec); }
-.feature-icon { display: grid; place-items: center; width: 42px; height: 42px; margin-bottom: 32px; border-radius: 12px; color: #08705e; background: #dbf2ea; font-size: 22px; }
-.blue .feature-icon { color: #3156b8; background: #e5ebff; }
-.amber .feature-icon { color: #9b6818; background: #f8eac8; }
-.feature-label { margin-bottom: 7px; color: #718078; font: 700 10px/1.3 ui-monospace, SFMono-Regular, Menlo, monospace; letter-spacing: 0.08em; text-transform: uppercase; }
-.feature-card strong { font-size: 20px; }
-.feature-description { margin-top: 10px; color: #66736c; font-size: 13px; line-height: 1.6; }
-.feature-arrow { position: absolute; right: 22px; bottom: 18px; color: #6a7870; font-size: 19px; }
-.benchmark-callout { display: flex; align-items: end; justify-content: space-between; gap: 35px; margin-top: 62px; padding: 36px; border-radius: 17px; color: white; background: linear-gradient(125deg, #142a22, #1c473b); }
-.benchmark-callout .section-kicker { color: #74d4bb; }
-.benchmark-callout h2 { color: white; }
-.benchmark-callout p:not(.section-kicker) { max-width: 610px; margin: 10px 0 0; color: #c3d4cd; line-height: 1.65; }
-.benchmark-callout > a { flex: none; padding: 10px 14px; border: 1px solid rgba(255,255,255,.25); border-radius: 9px; color: white; text-decoration: none; font-size: 13px; font-weight: 700; }
-.article-meta { display: flex; gap: 8px; margin-bottom: 22px; color: #7d8d84; font: 600 11px/1.4 ui-monospace, SFMono-Regular, Menlo, monospace; }
-.article-meta span:first-child { padding: 3px 8px; border-radius: 999px; color: #08705e; background: #e6f5ef; }
-.markdown-body { color: #27332d; font-size: 16px; line-height: 1.78; }
-.markdown-body > :first-child { margin-top: 0; }
-.markdown-body h1 { margin: 0 0 30px; color: #112019; font-size: clamp(36px, 5vw, 54px); line-height: 1.06; letter-spacing: -0.04em; }
-.markdown-body h2 { margin: 58px 0 18px; padding-top: 4px; color: #14241c; font-size: 27px; line-height: 1.25; letter-spacing: -0.02em; }
-.markdown-body h3 { margin: 35px 0 12px; color: #1f3027; font-size: 20px; line-height: 1.35; }
-.markdown-body p, .markdown-body ul, .markdown-body ol { margin: 14px 0; }
-.markdown-body li { margin: 5px 0; }
-.markdown-body a { color: var(--green); text-decoration-color: #9cc8ae; text-underline-offset: 3px; }
-.markdown-body a:hover { text-decoration-color: var(--green); }
-.markdown-body img { display: block; max-width: 100%; height: auto; margin: 30px auto; border-radius: 14px; }
-/* README badges are linked images. Keep them in a compact row instead of applying the figure layout above. */
-.article[data-group="overview"] .markdown-body > p:has(> a > img[src*="img.shields.io"]) {
-  display: flex;
-  flex-wrap: wrap;
-  justify-content: center;
-  gap: 7px;
-  max-width: 690px;
-  margin: 20px auto;
-}
-.article[data-group="overview"] .markdown-body > p > a > img[src*="img.shields.io"] {
-  display: inline-block;
-  width: auto;
-  max-height: 24px;
-  margin: 0;
-  border-radius: 4px;
-  vertical-align: middle;
-}
-.markdown-body p[align="center"] { text-align: center; }
-.article[data-group="overview"] .markdown-body > p:first-child img { max-width: min(470px, 72%); margin-top: 8px; filter: drop-shadow(0 18px 25px rgba(44, 85, 164, 0.1)); }
-.markdown-body code { padding: 2px 5px; border-radius: 5px; background: var(--code); color: #315844; font: 0.88em/1.5 ui-monospace, SFMono-Regular, Menlo, Consolas, monospace; }
-.markdown-body pre { margin: 24px 0; padding: 19px 21px; overflow-x: auto; border: 1px solid #dce5e0; border-radius: 12px; background: linear-gradient(145deg, #f5f8f6, #f8faf9); box-shadow: inset 3px 0 0 #7bd0b6; }
-.markdown-body pre code { padding: 0; color: #263b30; background: none; font-size: 13px; }
-.markdown-body blockquote { margin: 24px 0; padding: 4px 20px; border-left: 3px solid #68aa85; color: #53635a; background: #f5faf7; }
-.markdown-body table { display: block; width: 100%; margin: 24px 0; overflow-x: auto; border-collapse: collapse; font-size: 14px; }
-.markdown-body th, .markdown-body td { padding: 10px 13px; border: 1px solid var(--line); text-align: left; }
-.markdown-body th { background: #f4f6f3; }
-.markdown-body hr { margin: 44px 0; border: 0; border-top: 1px solid var(--line); }
-.article-footer { margin-top: 70px; padding-top: 20px; border-top: 1px solid var(--line); }
-.article-footer a { color: var(--muted); text-decoration: none; font-size: 13px; }
-.article-footer a:hover { color: var(--green); }
-.loading-line { width: 55%; height: 12px; margin-top: 40px; border-radius: 9px; background: linear-gradient(90deg, #edf0ec, #f8faf7, #edf0ec); background-size: 200% 100%; animation: loading 1.2s infinite; }
-@keyframes loading { to { background-position: -200% 0; } }
-
-.toc-panel { position: sticky; top: 66px; height: calc(100vh - 66px); padding: 58px 24px; overflow-y: auto; border-left: 1px solid #e9efeb; background: rgba(252, 253, 252, 0.92); }
-.toc h2 { margin: 0 0 12px; color: #8a948e; font-size: 10px; letter-spacing: 0.1em; text-transform: uppercase; }
-.toc a { display: block; padding: 5px 0; color: #7a857e; text-decoration: none; font-size: 12px; line-height: 1.45; }
-.toc a:hover { color: var(--green); }
-.toc .toc-h3 { padding-left: 12px; }
-.sidebar-backdrop { display: none; }
-
-@media (max-width: 1120px) {
-  .docs-shell { grid-template-columns: 250px minmax(0, 1fr); }
-  .toc-panel { display: none; }
-}
-
-@media (max-width: 760px) {
-  .topbar { height: 60px; padding: 0 16px; }
-  .brand-section, .brand-divider, .github-link { display: none; }
-  .quick-start-link { font-size: 12px; }
-  .menu-button { display: block; max-width: 112px; overflow: hidden; text-overflow: ellipsis; white-space: nowrap; }
-  .top-actions { gap: 9px; }
-  .docs-shell { display: block; padding-top: 60px; }
-  .sidebar { position: fixed; z-index: 25; top: 60px; bottom: 0; left: 0; width: min(310px, 86vw); height: auto; transform: translateX(-105%); transition: transform 180ms ease; box-shadow: 16px 0 35px rgba(20, 40, 30, 0.12); }
-  .sidebar.open { transform: translateX(0); }
-  .sidebar-backdrop { position: fixed; z-index: 20; inset: 60px 0 0; width: 100%; border: 0; background: rgba(18, 30, 23, 0.35); }
-  .sidebar-backdrop.visible { display: block; }
-  .article-wrap { padding: 38px 20px 72px; }
-  .article-meta { overflow: hidden; white-space: nowrap; text-overflow: ellipsis; }
-  .markdown-body { font-size: 15px; }
-  .markdown-body h1 { font-size: 34px; }
-  .markdown-body h2 { margin-top: 46px; font-size: 24px; }
-  .article[data-group="overview"] .markdown-body > p:first-child img { max-width: 84%; }
-  .home-hero { padding: 20px 0 54px; }
-  .home-hero h1 { font-size: 43px; }
-  .home-lead { font-size: 16px; }
-  .feature-grid { grid-template-columns: 1fr; }
-  .feature-card { min-height: 225px; }
-  .benchmark-callout { align-items: flex-start; flex-direction: column; padding: 28px 24px; }
-}
--- a/github-pages/tests/content-manifest.test.mjs
+++ b/github-pages/tests/content-manifest.test.mjs
@ -1,23 +0,0 @@
-import assert from "node:assert/strict";
-import { readFile } from "node:fs/promises";
-import test from "node:test";
-
-const manifestUrl = new URL("../.generated/content/manifest.json", import.meta.url);
-
-test("omits retired Agent documents and places Agent integration after getting started", async () => {
-  const manifest = JSON.parse(await readFile(manifestUrl, "utf8"));
-  const documents = manifest.documents;
-  const groups = [...new Set(documents.map((document) => document.group))];
-
-  assert.equal(documents.some((document) => document.id === "reme-memory-skill"), false);
-  assert.equal(documents.some((document) => document.sourcePath.endsWith("agent_integration_plan.md")), false);
-  assert.deepEqual(
-    documents
-      .filter((document) => document.group === "plugins")
-      .map((document) => document.title || document.titles?.en),
-    ["每日论文插件", "Auto Fin 插件", "Daily Paper Plugin", "Auto Fin Plugin"],
-  );
-  assert.equal(documents.some((document) => document.group === "cookbooks"), false);
-  assert.ok(groups.indexOf("integration") > groups.indexOf("start"));
-  assert.ok(groups.indexOf("integration") < groups.indexOf("fundamentals"));
-});
--- a/github-pages/tests/generated-content.test.mjs
+++ b/github-pages/tests/generated-content.test.mjs
@ -0,0 +1,90 @@
+import assert from "node:assert/strict";
+import { access, readFile } from "node:fs/promises";
+import path from "node:path";
+import test from "node:test";
+import { fileURLToPath } from "node:url";
+import { parse as parseYaml } from "yaml";
+import { legacyRoutes } from "../../docs/.vitepress/legacy-routes.mjs";
+
+const siteDir = path.resolve(path.dirname(fileURLToPath(import.meta.url)), "..");
+const repoDir = path.resolve(siteDir, "..");
+const generatedDir = path.join(siteDir, ".generated", "site");
+
+test("generates every required bilingual guide", async () => {
+  const names = [
+    "configuration.md",
+    "services.md",
+    "operations.md",
+    "integrations.md",
+    "plugin_development.md",
+    "faq.md",
+    "reference/cli.md",
+    "reference/jobs.md",
+  ];
+  for (const language of ["zh", "en"]) {
+    for (const name of names) await access(path.join(generatedDir, language, name));
+  }
+  await access(path.join(generatedDir, "zh/integrations/claude-code.md"));
+  await access(path.join(generatedDir, "en/integrations/claude-code.md"));
+  await access(path.join(generatedDir, "zh/integrations/hermes.md"));
+  await access(path.join(generatedDir, "en/integrations/hermes.md"));
+});
+
+test("maps mirrored pages back to their canonical repository sources", async () => {
+  const sourceMap = JSON.parse(await readFile(path.join(generatedDir, ".source-map.json"), "utf8"));
+  assert.equal(sourceMap["zh/integrations/typescript.md"], "typescript/README_ZH.md");
+  assert.equal(sourceMap["en/integrations/claude-code.md"], "integrations/claude_code/README.md");
+  assert.equal(sourceMap["en/integrations/hermes.md"], "integrations/hermes_agent/README.md");
+  assert.equal(sourceMap["en/workspace/studio.md"], "reme_studio/README.md");
+  assert.equal(sourceMap["zh/plugins/lme.md"], "plugins/lme/README_ZH.md");
+  assert.equal(sourceMap["en/reference/jobs.md"], "reme/config/default.yaml");
+});
+
+test("generates the callable Job reference from default.yaml", async () => {
+  const config = parseYaml(await readFile(path.join(repoDir, "reme/config/default.yaml"), "utf8"));
+  const callableJobs = Object.entries(config.jobs)
+    .filter(([, job]) => !["background", "cron"].includes(job.backend))
+    .map(([name]) => name);
+
+  for (const language of ["zh", "en"]) {
+    const reference = await readFile(path.join(generatedDir, language, "reference", "jobs.md"), "utf8");
+    for (const job of callableJobs) assert.ok(reference.includes(`### ${"`"}${job}${"`"}`), job);
+  }
+});
+
+test("keeps generated content disposable and excludes internal plans", async () => {
+  await assert.rejects(access(path.join(generatedDir, "plans")));
+  await access(path.join(generatedDir, ".vitepress", "config.mts"));
+  await access(path.join(generatedDir, "public", "favicon.svg"));
+  assert.equal(
+    (await readFile(path.join(generatedDir, "public", "CNAME"), "utf8")).trim(),
+    "reme.agentscope.io",
+  );
+});
+
+test("maps every legacy query-string document ID to a generated page", async () => {
+  assert.equal(Object.keys(legacyRoutes).length, 45);
+  assert.equal(legacyRoutes["studio-en"], "/en/workspace/studio");
+  assert.equal(legacyRoutes["en-quick_start"], "/en/quick_start");
+  assert.equal(legacyRoutes["agents-guide"], "https://github.com/agentscope-ai/ReMe/blob/main/AGENTS.md");
+
+  for (const [id, route] of Object.entries(legacyRoutes)) {
+    if (route.startsWith("https://")) continue;
+    assert.match(route, /^\/(?:zh|en)\//, id);
+    const relative = route.endsWith("/") ? `${route.slice(1)}index.md` : `${route.slice(1)}.md`;
+    await access(path.join(generatedDir, relative));
+  }
+});
+
+test("tracks every generated input in documentation CI and deployment", async () => {
+  const requiredPaths = [
+    "reme/config/default.yaml",
+    "integrations/claude_code/README.md",
+    "integrations/hermes_agent/README.md",
+    "benchmark/toolmemory/gitcha.png",
+  ];
+  for (const workflow of ["ci-docs.yml", "deploy-docs.yml"]) {
+    const source = await readFile(path.join(repoDir, ".github/workflows", workflow), "utf8");
+    for (const requiredPath of requiredPaths) assert.ok(source.includes(requiredPath), `${workflow}: ${requiredPath}`);
+  }
+});
--- a/github-pages/tests/markdown.test.mjs
+++ b/github-pages/tests/markdown.test.mjs
@ -1,15 +0,0 @@
-import assert from "node:assert/strict";
-import test from "node:test";
-import { stripMarkdownFrontmatter } from "../src/markdown.js";
-
-test("strips leading YAML frontmatter before rendering", () => {
-  assert.equal(
-    stripMarkdownFrontmatter("---\nname: reme_memory\ndescription: Memory skill\n---\n\n# ReMe Memory\n"),
-    "\n# ReMe Memory\n",
-  );
-});
-
-test("preserves Markdown without frontmatter", () => {
-  const markdown = "# ReMe Memory\n\nContent\n";
-  assert.equal(stripMarkdownFrontmatter(markdown), markdown);
-});
--- a/github-pages/vite.config.js
+++ b/github-pages/vite.config.js
@ -1,10 +0,0 @@
-import { defineConfig } from "vite";
-
-export default defineConfig({
-  base: "./",
-  publicDir: ".generated",
-  build: {
-    outDir: "dist",
-    emptyOutDir: true,
-  },
-});