Media Crawler Easy

多平台社媒评论数据采集爬虫工具。覆盖 B站(bilibili)/小红书/抖音(douyin)/快手/微博/贴吧/知乎 7大平台。当用户提到"爬评论""采集评论""下载评论""爬数据""批量采集""舆情分析""评论分析""社媒数据" "抓取评论""导出评论""获取评论""视频评论""帖子评论""笔记评论""UP主评论""博主评论" "关键词搜索爬取""社交媒体数据采集""爬虫""数据爬取""评论数据""用户评论"时自动触发。支持按关键词搜索、指定链接采集、创作者主页三种模式。生成交互式 HTML 报告 + 数据文件。

15 starsData & Analytics

Install

skills CLI (Claude Code, Cursor, Codex and more)

npx skills add isheng-eqi/media-crawler-easy --skill media-crawler-easy

Manual install

git clone --depth 1 https://github.com/isheng-eqi/media-crawler-easy.git /tmp/media-crawler-easy
mkdir -p ~/.claude/skills
cp -r /tmp/media-crawler-easy/. ~/.claude/skills/media-crawler-easy

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# MediaCrawler 多平台采集助手

基于 [MediaCrawler](https://github.com/NanmiCoder/MediaCrawler) (上游 65K+ star)，7 大平台社媒数据采集。

> 目录约定：下文 `<skill>` 指本 SKILL.md 所在的目录。
> 插件方式安装时它等于 `${CLAUDE_PLUGIN_ROOT}`；手动 clone 安装时它等于 `~/.claude/skills/media-crawler-easy`。
> 脚本内部路径是相对自身解析的，不依赖具体安装位置。

## ⚠️ 关键约束

1. **严格 3 轮问答**，每轮只问一个问题，不提前列出所有选项。
2. **参考材料仅排查时查阅**，不主动展示给用户。
3. **每次不超过 4 个选项**。
4. **环境检测只报结论**："X 项通过，Y 项修复，可以继续"。

## 执行流程

```
① 配置问答（3轮）→ ② 环境搭建 → ③ 写入配置 →
④ 执行爬取 → ⑤ 生成报告+验证 → ⑥ 汇报+清理
```

---

## ① 配置问答（严格 3 轮，不可合并）

### 第 1 轮：只问平台

```
要爬哪个平台？
B站(bili) / 小红书(xhs) / 抖音(dy) / 快手(ks) / 微博(wb) / 贴吧(tieba) / 知乎(zhihu)
```

用户回答后再问模式：

```
用什么方式爬？
A. 关键词搜索  B. 指定链接  C. 创作者主页
```

### 第 2 轮：只问具体参数

根据第 1 轮回答：
- **选 A**：问「关键词是什么？每个关键词爬多少条？（默认20）」
- **选 B**：问「把链接发给我」
- **选 C**：问「把创作者主页链接或 UID 发给我」

### 第 3 轮：只问评论设置

```
每条内容爬多少评论？（默认30）
要楼中楼吗？（默认不要）
要词云图吗？（默认不要）
爬取间隔设几秒？（默认2秒）
```

用户回复"默认"即全部用默认值。

---

## ② 环境搭建

```bash
cd <skill> && python scripts/env_setup.py
```

报结论即可。有 FAIL 先修再继续；不确定错误查 `references/env-troubleshooting.md`。

---

## ③ 写入配置

Edit 修改 `config/base_config.py`：`PLATFORM`, `KEYWORDS`, `CRAWLER_TYPE`, `CRAWLER_MAX_NOTES_COUNT`, `CRAWLER_MAX_COMMENTS_COUNT_SINGLENOTES`, `ENABLE_GET_SUB_COMMENTS`, `ENABLE_GET_WORDCLOUD`, `CRAWLER_MAX_SLEEP_SEC`。

平台专属 config 只改当前平台对应的文件（如抖音改 `dy_config.py`）。

**重要**：`AUTO_CLOSE_BROWSER` 默认为 `False`（保持浏览器打开）。**永远不要把它改成 True**——否则爬虫结束/出错时 Chrome 会立刻关闭，用户看不到状态。

---

## ④ 执行爬取

用 `run_in_background: true` 启动，每 20-30 秒检查日志。

Read the full skill on GitHub

Details

Skill name: media-crawler-easy
Repository: isheng-eqi/media-crawler-easy
License: MIT
Repo updated: 25 September 2026

FAQ

What does the Media Crawler Easy skill do?

How do I install the Media Crawler Easy skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add isheng-eqi/media-crawler-easy --skill media-crawler-easy. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Media Crawler Easy popular?

It lives in isheng-eqi/media-crawler-easy, which has 15 GitHub stars and 0 forks.

Related skills

Clawarr Suite

omiron33/clawarr-suite

Comprehensive management for self-hosted media stacks (Sonarr, Radarr, Lidarr, Readarr, Prowlarr, Bazarr, Overseerr, Plex, Tautulli, SABnzbd, Recyclarr, Unpackerr, Notifiarr, Maintainerr, Kometa, FlareSolverr). Deep library exploration, analytics, dashboard generation, content management, request handling, subtitle management, indexer control, download monitoring, quality profile sync, library cleanup automation, notification routing, collection/overlay management, and media tracker integration (Trakt, Letterboxd, Simkl).

Data & Analytics

Aiz Infographic

aizzaku/aiz-infographic

An infographic skill for AI agents — works inside Claude Code, Codex, Hermes, OpenClaw, Cursor, and any agent host that loads skills. Create professional infographics as self-contained HTML files with PNG exports. Use when the user asks to create, design, generate, build, or make an infographic, data visualization, visual explainer, chart poster, cheatsheet, or pocket guide. Trigger phrases include "make this visual", "turn this into a graphic", "visualize this data", "infographic for X", "cheatsheet for X", or when the user provides structured data (tokenomics allocations, ecosystem/partner l

Data & Analytics

Notebooklm Course Builder

hmj1026/notebooklm-course-builder

以條件式瀏覽器自動化或 14 階段嚮導建構 NotebookLM 課程教材。適用於：(1) 將課程大綱轉為包含來源引用、逐節講義、Quiz 與 Study Guide 的完整課程；(2) 規劃 Notebook 與研究分群、來源審核、Coverage 分析；(3) 自動操作 NotebookLM 或延續既有建置進度。

Data & Analytics

Ziwei Doushu

Linden-TR/ziwei-doushu-skill

When the user asks about 紫微斗数 / Ziwei Doushu / Purple Star Astrology — including 排盘 (chart casting), 命盘解读 (chart reading), 看运势 (fortune reading), 事业/感情/财运/健康等命理咨询, or mentions 十四主星 / 十二宫 / 四化 / 大限 / 流年 / 飞星 / 七星 / 六吉六煞 — use this skill. Also trigger when the user says 算命, 看盘, 算紫微, 帮我看运势, or any request for Chinese metaphysical destiny analysis. Multi-school methodology: Zhongzhou (primary), San He, Fei Xing, Qin Tian Si Hua, He Luo.

Data & Analytics

Hkdata

gnokit/hkdata

Use when the user asks about Hong Kong government open data, statistics, or datasets — e.g. transport, weather, population, employment, finance, health, education, housing, crime, tourism, facilities, etc. — in English or Chinese. Queries data.gov.hk and returns sourced, live answers, or an honest "no dataset".

Data & Analytics

Lemonwrench

bpawnzZ/lemonwrench

Use when looking up car repair procedures, torque specs, diagrams or service-manual steps. Scrapes free OEM manuals by vehicle, resolves the exact trim/engine, and cross-checks parts catalogs and field sources.

Data & Analytics

Browse the most popular Claude skills, all data & analytics skills, or write your own with the skills generator.