feat(linux-do): split topic content into a dedicated command

README.zh-CN.md

@@ -178,7 +178,7 @@ npx skills add jackwener/opencli --skill opencli-oneshot     # 快速命令参
 | **jike** | `feed` `search` `create` `like` `comment` `repost` `notifications` `post` `topic` `user` | 浏览器 |
 | **jimeng** | `generate` `history` | 浏览器 |
 | **yollomi** | `generate` `video` `edit` `upload` `models` `remove-bg` `upscale` `face-swap` `restore` `try-on` `background` `object-remover` | 浏览器 |
-| **linux-do** | `hot` `latest` `feed` `search` `categories` `category` `tags` `topic` `user-posts` `user-topics` | 浏览器 |
+| **linux-do** | `hot` `latest` `feed` `search` `categories` `category` `tags` `topic` `topic-content` `user-posts` `user-topics` | 浏览器 |
 | **stackoverflow** | `hot` `search` `bounties` `unanswered` | 公开 |
 | **steam** | `top-sellers` | 公开 |
 | **weread** | `shelf` `search` `book` `highlights` `notes` `notebooks` `ranking` | 浏览器 |

clis/linux-do/topic-content.test.ts

@@ -0,0 +1,67 @@
+import { getRegistry } from '@jackwener/opencli/registry';
+import fs from 'node:fs';
+import { describe, expect, it } from 'vitest';
+import { __test__ } from './topic-content.js';
+
+describe('linux-do topic-content', () => {
+  it('prefers raw markdown when the topic payload includes it', () => {
+    const result = __test__.extractTopicContent({
+      title: 'Hello Linux.do',
+      post_stream: {
+        posts: [
+          {
+            post_number: 1,
+            username: 'neo',
+            raw: '## Heading\n\n- one\n- two',
+            cooked: '<h2>Heading</h2><ul><li>one</li><li>two</li></ul>',
+            like_count: 7,
+            created_at: '2025-04-05T10:00:00.000Z',
+          },
+        ],
+      },
+    }, 1234);
+
+    expect(result.content).toContain('---');
+    expect(result.content).toContain('title: Hello Linux.do');
+    expect(result.content).toContain('author: neo');
+    expect(result.content).toContain('likes: 7');
+    expect(result.content).toContain('url: https://linux.do/t/1234');
+    expect(result.content).toContain('## Heading');
+    expect(result.content).toContain('- one');
+  });
+
+  it('falls back to cooked html and converts it to markdown', () => {
+    const result = __test__.extractTopicContent({
+      title: 'Converted Topic',
+      post_stream: {
+        posts: [
+          {
+            post_number: 1,
+            username: 'trinity',
+            cooked: '<p>Hello <strong>world</strong></p><blockquote><p>quoted</p></blockquote>',
+            like_count: 3,
+            created_at: '2025-04-05T10:00:00.000Z',
+          },
+        ],
+      },
+    }, 42);
+
+    expect(result.content).toContain('Hello **world**');
+    expect(result.content).toContain('> quoted');
+  });
+
+  it('registers topic-content with plain default output for markdown body rendering', () => {
+    const command = getRegistry().get('linux-do/topic-content');
+
+    expect(command?.defaultFormat).toBe('plain');
+    expect(command?.columns).toEqual(['content']);
+  });
+
+  it('keeps topic yaml as a summarized first-page reader after the split', () => {
+    const topicYaml = fs.readFileSync(new URL('./topic.yaml', import.meta.url), 'utf8');
+
+    expect(topicYaml).not.toContain('main_only');
+    expect(topicYaml).toContain('slice(0, 200)');
+    expect(topicYaml).toContain('帖子首页摘要和回复');
+  });
+});

clis/linux-do/topic-content.ts

@@ -0,0 +1,207 @@
+import { AuthRequiredError, CommandExecutionError, EmptyResultError } from '@jackwener/opencli/errors';
+import { cli, Strategy } from '@jackwener/opencli/registry';
+import type { IPage } from '@jackwener/opencli/types';
+import { htmlToMarkdown, isRecord } from '@jackwener/opencli/utils';
+const LINUX_DO_DOMAIN = 'linux.do';
+const LINUX_DO_HOME = 'https://linux.do';
+
+interface FetchTopicResult {
+  ok: boolean;
+  status?: number;
+  data?: unknown;
+  error?: string;
+}
+
+interface LinuxDoTopicPost {
+  post_number?: number;
+  username?: string;
+  raw?: string;
+  cooked?: string;
+  like_count?: number;
+  created_at?: string;
+}
+
+interface LinuxDoTopicPayload {
+  title?: string;
+  post_stream?: {
+    posts?: LinuxDoTopicPost[];
+  };
+}
+
+interface TopicContentRow {
+  content: string;
+}
+
+function toLocalTime(utcStr: string): string {
+  if (!utcStr) return '';
+  const date = new Date(utcStr);
+  return Number.isNaN(date.getTime()) ? utcStr : date.toLocaleString();
+}
+
+function normalizeTopicPayload(payload: unknown): LinuxDoTopicPayload | null {
+  if (!isRecord(payload)) return null;
+  const postStream = isRecord(payload.post_stream)
+    ? {
+      posts: Array.isArray(payload.post_stream.posts)
+        ? payload.post_stream.posts.filter(isRecord).map((post) => ({
+          post_number: typeof post.post_number === 'number' ? post.post_number : undefined,
+          username: typeof post.username === 'string' ? post.username : undefined,
+          raw: typeof post.raw === 'string' ? post.raw : undefined,
+          cooked: typeof post.cooked === 'string' ? post.cooked : undefined,
+          like_count: typeof post.like_count === 'number' ? post.like_count : undefined,
+          created_at: typeof post.created_at === 'string' ? post.created_at : undefined,
+        }))
+        : undefined,
+    }
+    : undefined;
+
+  return {
+    title: typeof payload.title === 'string' ? payload.title : undefined,
+    post_stream: postStream,
+  };
+}
+
+function buildTopicMarkdownDocument(params: {
+  title: string;
+  author: string;
+  likes?: number;
+  createdAt: string;
+  url: string;
+  body: string;
+}): string {
+  const frontMatterLines: string[] = [];
+  const entries: [string, string | number | undefined][] = [
+    ['title', params.title || undefined],
+    ['author', params.author || undefined],
+    ['likes', typeof params.likes === 'number' && Number.isFinite(params.likes) ? params.likes : undefined],
+    ['createdAt', params.createdAt || undefined],
+    ['url', params.url || undefined],
+  ];
+  for (const [key, value] of entries) {
+    if (value === undefined) continue;
+    if (typeof value === 'number') {
+      frontMatterLines.push(`${key}: ${value}`);
+    } else {
+      // Quote strings that could be misinterpreted by YAML parsers
+      const needsQuote = /[#{}[\],&*?|>!%@`'"]/.test(value) || /: /.test(value) || /:$/.test(value) || value.includes('\n');
+      frontMatterLines.push(`${key}: ${needsQuote ? `'${value.replace(/'/g, "''")}'` : value}`);
+    }
+  }
+  const frontMatter = frontMatterLines.join('\n');
+
+  return [
+    frontMatter ? `---\n${frontMatter}\n---` : '',
+    params.body.trim(),
+  ].filter(Boolean).join('\n\n').trim();
+}
+
+function extractTopicContent(payload: unknown, id: number): TopicContentRow {
+  const topic = normalizeTopicPayload(payload);
+  if (!topic) {
+    throw new CommandExecutionError('linux.do returned an unexpected topic payload');
+  }
+
+  const posts = topic.post_stream?.posts ?? [];
+  const mainPost = posts.find((post) => post.post_number === 1);
+  if (!mainPost) {
+    throw new EmptyResultError('linux-do/topic-content', `Could not find the main post for topic ${id}.`);
+  }
+
+  const body = typeof mainPost.raw === 'string' && mainPost.raw.trim()
+    ? mainPost.raw.trim()
+    : htmlToMarkdown(mainPost.cooked ?? '');
+
+  if (!body) {
+    throw new EmptyResultError('linux-do/topic-content', `Topic ${id} does not contain a readable main post body.`);
+  }
+
+  return {
+    content: buildTopicMarkdownDocument({
+      title: topic.title?.trim() ?? '',
+      author: mainPost.username?.trim() ?? '',
+      likes: typeof mainPost.like_count === 'number' ? mainPost.like_count : undefined,
+      createdAt: toLocalTime(mainPost.created_at ?? ''),
+      url: `${LINUX_DO_HOME}/t/${id}`,
+      body,
+    }),
+  };
+}
+
+async function fetchTopicPayload(page: IPage, id: number): Promise<unknown> {
+  const result = await page.evaluate(`(async () => {
+    try {
+      const res = await fetch('/t/${id}.json?include_raw=true', { credentials: 'include' });
+      let data = null;
+      try {
+        data = await res.json();
+      } catch (_error) {
+        data = null;
+      }
+      return {
+        ok: res.ok,
+        status: res.status,
+        data,
+        error: data === null ? 'Response is not valid JSON' : '',
+      };
+    } catch (error) {
+      return {
+        ok: false,
+        error: error instanceof Error ? error.message : String(error),
+      };
+    }
+  })()`) as FetchTopicResult | null;
+
+  if (!result) {
+    throw new CommandExecutionError('linux.do returned an empty browser response');
+  }
+
+  if (result.status === 401 || result.status === 403) {
+    throw new AuthRequiredError(LINUX_DO_DOMAIN, 'linux.do requires an active signed-in browser session');
+  }
+
+  if (result.error === 'Response is not valid JSON') {
+    throw new AuthRequiredError(LINUX_DO_DOMAIN, 'linux.do requires an active signed-in browser session');
+  }
+
+  if (!result.ok) {
+    throw new CommandExecutionError(
+      result.error || `linux.do request failed: HTTP ${result.status ?? 'unknown'}`,
+    );
+  }
+
+  if (result.error) {
+    throw new CommandExecutionError(result.error, 'Please verify your linux.do session is still valid');
+  }
+
+  return result.data;
+}
+
+cli({
+  site: 'linux-do',
+  name: 'topic-content',
+  description: 'Get the main topic body as Markdown',
+  domain: LINUX_DO_DOMAIN,
+  strategy: Strategy.COOKIE,
+  browser: true,
+  defaultFormat: 'plain',
+  args: [
+    { name: 'id', positional: true, type: 'int', required: true, help: 'Topic ID' },
+  ],
+  columns: ['content'],
+  func: async (page: IPage, kwargs) => {
+    const id = Number(kwargs.id);
+    if (!Number.isInteger(id) || id <= 0) {
+      throw new CommandExecutionError(`Invalid linux.do topic id: ${String(kwargs.id ?? '')}`);
+    }
+
+    const payload = await fetchTopicPayload(page, id);
+    return [extractTopicContent(payload, id)];
+  },
+});
+
+export const __test__ = {
+  buildTopicMarkdownDocument,
+  extractTopicContent,
+  normalizeTopicPayload,
+  toLocalTime,
+};

clis/linux-do/topic.yaml

@@ -1,6 +1,6 @@
 site: linux-do
 name: topic
-description: linux.do 帖子详情和回复（首页）
+description: linux.do 帖子首页摘要和回复（首屏）
 domain: linux.do
 strategy: cookie
 browser: true
@@ -15,17 +15,12 @@ args:
     type: int
     default: 20
     description: Number of posts
-  main_only:
-    type: bool
-    default: false
-    description: Only return the main post body without truncation
 
 pipeline:
   - navigate: https://linux.do
 
   - evaluate: |
       (async () => {
-        const mainOnly = ${{ args.main_only }};
         const toLocalTime = (utcStr) => {
           if (!utcStr) return '';
           const date = new Date(utcStr);
@@ -50,16 +45,6 @@ pipeline:
           .replace(/\s+/g, ' ')
           .trim();
         const posts = data?.post_stream?.posts || [];
-        if (mainOnly) {
-          const mainPost = posts.find(p => p.post_number === 1);
-          if (!mainPost) return [];
-          return [{
-            author: mainPost.username || '',
-            content: mainPost.cooked || '',
-            likes: mainPost.like_count || 0,
-            created_at: toLocalTime(mainPost.created_at),
-          }];
-        }
         return posts.slice(0, ${{ args.limit }}).map(p => ({
           author: p.username,
           content: strip(p.cooked).slice(0, 200),

docs/adapters/browser/linux-do.md

@@ -11,6 +11,7 @@
 | `opencli linux-do tags` | List popular tags |
 | `opencli linux-do search <query>` | Search topics |
 | `opencli linux-do topic <id>` | View topic posts |
+| `opencli linux-do topic-content <id>` | Read the main topic body as Markdown |
 | `opencli linux-do user-topics <username>` | Topics created by a user |
 | `opencli linux-do user-posts <username>` | Replies posted by a user |
 
@@ -147,19 +148,35 @@ Output columns: `rank`, `title`, `views`, `likes`, `replies`, `url`
 
 ## topic
 
-View posts within a topic (first page).
+View summarized first-page posts within a topic.
 
 ```bash
 opencli linux-do topic 1234
 opencli linux-do topic 1234 --limit 50
-opencli linux-do topic 1234 --main_only -f json | jq -r '.[0].content'
 ```
 
 Notes:
-- `--main_only` returns only the main post row and keeps the body untruncated
+- `content` is a plain-text summary extracted from each first-page post
+- Each summary is truncated to 200 characters
+- Use `opencli linux-do topic-content <id>` for the full main post body in Markdown
 
 Output columns: `author`, `content`, `likes`, `created_at`
 
+## topic-content
+
+Read the main topic body as Markdown.
+
+```bash
+opencli linux-do topic-content 1234
+opencli linux-do topic-content 1234 -f json
+```
+
+Notes:
+- Default output prints the Markdown body directly for copy/paste or piping into LLMs
+- Use `-f json` if you want a machine-readable wrapper
+
+Output columns: `content`
+
 ## user-topics
 
 List topics created by a user.

docs/adapters/index.md

@@ -28,7 +28,7 @@ Run `opencli list` for the live registry.
 | **[jike](./browser/jike)** | `feed` `search` `post` `topic` `user` `create` `comment` `like` `repost` `notifications` | 🔐 Browser |
 | **[jimeng](./browser/jimeng)** | `generate` `history` | 🔐 Browser |
 | **[yollomi](./browser/yollomi)** | `generate` `video` `edit` `upload` `models` `remove-bg` `upscale` `face-swap` `restore` `try-on` `background` `object-remover` | 🔐 Browser |
-| **[linux-do](./browser/linux-do)** | `hot` `latest` `feed` `search` `categories` `category` `tags` `topic` `user-posts` `user-topics` | 🔐 Browser |
+| **[linux-do](./browser/linux-do)** | `hot` `latest` `feed` `search` `categories` `category` `tags` `topic` `topic-content` `user-posts` `user-topics` | 🔐 Browser |
 | **[chaoxing](./browser/chaoxing)** | `assignments` `exams` | 🔐 Browser |
 | **[grok](./browser/grok)** | `ask` | 🔐 Browser |
 | **[gemini](./browser/gemini)** | `new` `ask` `image` `deep-research` `deep-research-result` | 🔐 Browser |

skills/opencli-usage/SKILL.md

@@ -99,7 +99,7 @@ Usage: `opencli <site> <command> [args] [--limit N] [-f json|yaml|md|csv|table]`
 | **web** | `read` — any URL to Markdown |
 | **weixin** | `download` — 公众号 article to Markdown |
 | **v2ex** (browser) | `daily` `me` `notifications` |
-| **linux-do** (browser) | `hot` `latest` `feed` `search` `categories` `category` `tags` `topic` `user-posts` `user-topics` |
+| **linux-do** (browser) | `hot` `latest` `feed` `search` `categories` `category` `tags` `topic` `topic-content` `user-posts` `user-topics` |
 | **bloomberg** (browser) | `news` — full article reader |
 | **grok** | `ask` |
 | **doubao** | `status` `new` `send` `read` `ask` `detail` `history` `meeting-summary` `meeting-transcript` |