yuzhihui886

Novel Scraper

Data Analysisv1.6.0
By yuzhihui886Updated 2026-05-119 versions · latest released 2026-04-04novel-scraper@clawhub_yuzhihui886/novel-scraper
Web Scraping#批量

According to AlphaMoat, as of 2026-09, Novel Scraper has 1,924 downloads and 5 stars — ranked #4,477 of 5,078 AI skills overall, and #348 of 399 in Data Analysis.

Installs
225
Downloads
1.9K
Stars
5
Overall Rank
#4,477
Top 89% overall · Top 88% in Data Analysis
PlatformInstallsDownloadsStarsComments
ClawHub ↗391.4K4-
SkillHub ↗2251.9K5—

Platforms count differently and update at different times.

What this skill does

By default every 10 chapters are merged into one document to avoid scattered files. The smart V5 edition auto-detects and completes pagination (up to 5 pages), parses chapter numbers from titles, intelligently skips non-novel content such as author's notes and lottery notices, and runs basic quality verification and continuity checks; the v1.4.0 production edition infers chapter numbers from URL IDs and is the fastest, suited to daily batch scraping of consecutive known URLs. The CLI supports --chapters 301-400 for precise chapter-range selection (v1.5.0 fixed an index-slicing bug), --urls for multiple URLs, --book (auto-extracted by default), --merge-interval (default 10), --memory-limit (default 2500MB), --auto-close (default every 3 chapters), and --retry (default 3). Output goes to ~/.openclaw/workspace/novels/ with names like 书名_第 X-Y 章.txt. The scraping makes outbound web requests and follows redirects, so it should be used only with trusted public novel-site URLs, and that fetched text, logs, cache, catalog, and progress data are stored locally.

Profile published 2026-10-29

What you can do with it

  • Batch-scrape web novels by chapter range
  • Complete paginated chapters and skip non-novel content automatically
  • Merge output into TXT files (10 chapters per document by default)

When to use it

  • When web-novel chapters need scraping, bulk downloading, or saving as TXT
  • When chapters are paginated or precise chapter-number selection is needed

Example prompts

Generated from the skill’s documentation as examples; not results of actual runs.

Scrape chapters 301-400 of 《没钱修什么仙》
Use the smart V5 edition to scrape paginated chapters with auto completion
Save chapters 111-210 merged into TXT files of 10 chapters each

Monthly Trend

Downloads1.9K
5321.3K2.0K2026-03 · Downloads 6352026-05 · Downloads 1.2K2026-06 · Downloads 1.3K2026-07 · Downloads 1.5K2026-08 · Downloads 1.6K2026-09 · Downloads 1.9K26-0326-09
Stars5
3452026-03 · Stars 32026-05 · Stars 52026-06 · Stars 52026-07 · Stars 52026-08 · Stars 52026-09 · Stars 526-0326-09

The counting method changed in 2026-09: skills with the same name are now counted separately by author, and the full catalog is covered. Dashed lines show months before the change; they are not compared with later months, and no growth is computed across the change.

MonthDownloadsMoMStars
2026-091.9K—5
2026-081.6K+7.3%5
2026-071.5K+16.1%5
2026-061.3K+9.4%5
2026-051.2K—5
2026-03635—3

SKILL.md (original)

The author’s original file, shown as plain text without edits or review by AlphaMoat. Links are not clickable; check commands before running them. Fetched 2026-10-23.

4.2 KB · ClawHub v1.6.0
---
name: novel-scraper
description: >-
  智能小说抓取工具,支持自动翻页、分页补全、章节号自动解析。
  使用 curl+BeautifulSoup 抓取笔趣阁等小说网站,输出格式化 TXT 文件。
  默认每 10 章合并为一个文档,避免文件零散分布。
  自动检测分页并补全,智能跳过非小说内容(作者感言、抽奖预告等)。
  Use when: 抓取网络小说章节、批量下载小说内容、保存小说为 TXT 格式。
---
Show all (165 lines)
# Novel Scraper - 小说抓取工具

**版本说明**:novel-scraper 提供两个版本,根据需求选择使用。

---

## 📦 版本对比

| 特性 | v1.4.0(生产版) | V5(智能版) |
|------|------------------|--------------|
| **章节号解析** | URL ID 推算 | 标题提取 ✅ |
| **分页检测** | ❌ 无 | ✅ 自动 ✅ |
| **分页补全** | ❌ 无 | ✅ 自动(最多 5 页)✅ |
| **非小说跳过** | ❌ 无 | ✅ 智能判断 ✅ |
| **质量验证** | ❌ 无 | ✅ 基础验证 ✅ |
| **连续性检查** | ❌ 无 | ✅ 自动检查 ✅ |
| **代码量** | 34KB | 18KB |
| **速度** | ⚡ 最快 | ⚡ 快 |
| **推荐场景** | 日常批量抓取 | 高质量/分页章节 |

---

## 🚀 快速开始

### v1.5.0(推荐)- 按章节号抓取

```bash
cd ~/.openclaw/workspace/skills/novel-scraper
python3 scripts/scraper_v5.py \
  --chapters 301-400 \
  --book "没钱修什么仙" \
  --merge-interval 10
```

**特点**:
- ✅ **按章节号精确筛选**(修复索引切片 bug)
- ✅ 自动提示缺失章节
- ✅ 分页自动补全
- ✅ 适合按章节范围抓取的场景

---

### v1.4.0(生产版)- 日常使用

```bash
cd ~/.openclaw/workspace/skills/novel-scraper
python3 scripts/scraper.py \
  --urls "https://www.bqquge.com/4/1962,https://www.bqquge.com/4/1963,..." \
  --book "没钱修什么仙" \
  --merge-interval 10
```

**特点**:
- ✅ 稳定可靠
- ✅ 速度最快
- ✅ 适合已知章节 URL 连续的场景

---

### V5(智能版)- 高质量抓取

```bash
python3 scripts/scraper_v5.py \
  --chapters 301-400 \
  --book "没钱修什么仙" \
  --merge-interval 10
```

**特点**:
- ✅ 章节号自动解析
- ✅ 分页自动补全
- ✅ 非小说内容智能跳过
- ✅ 适合分页章节较多的场景

---

## 📋 命令行参数

### v1.5.0 参数(推荐)

| 参数 | 说明 | 默认值 |
|------|------|--------|
| `--chapters` | **章节号范围**(格式:`起始 - 结束`) | - |
| `--url` | 单章 URL | - |
| `--urls` | 多章 URL(逗号分隔) | - |
| `--book` | 书名 | 自动提取 |
| `--merge-interval` | 每 N 章合并 | 10 |
| `--memory-limit` | 内存限制 MB | 2500 |
| `--auto-close` | 每 N 章释放内存 | 3 |
| `--retry` | 重试次数 | 3 |

**⚠️ v1.5.0 重要修复**:
- 修复索引切片 bug(之前 `data[300:400]` 错误对应章节号 300-419)
- 新增 `--chapters` 参数,按章节号精确筛选
- 自动提示缺失章节(网站目录中没有的章节)

### V5 参数

| 参数 | 说明 | 默认值 |
|------|------|--------|
| `--chapters` | **章节号范围**(格式:`起始 - 结束`) | - |
| `--url` | 单章 URL | - |
| `--urls` | 多章 URL(逗号分隔) | - |
| `--book` | 书名 | 自动提取 |
| `--merge-interval` | 每 N 章合并 | 10 |
| `--strict` | 严格质量验证 | 关闭 |
| `-v` | 详细日志 | 关闭 |

---

## 📁 输出位置

```
~/.openclaw/workspace/novels/
```

**文件名格式**:
- 多章合并:`书名_第 X-Y 章.txt`
- 单章:`书名_第 X 章.txt`

---

## 🧪 测试验证

| 测试场景 | 版本 | 结果 |
|----------|------|------|
| 第 1-10 章 | V5 | ✅ 119KB,完整 |
| 第 11-20 章 | V5 | ✅ 100KB,完整 |
| 第 111-210 章 | v1.4.0 | ✅ 10 个文件,~700KB |
| 分页补全 | V5 | ✅ 28 段→112 段 |
| 非小说跳过 | V5 | ✅ 自动跳过 |

---

## 🔧 故障排除

### 抓取内容为空
1. 检查 URL 格式
2. 清除缓存:`rm -rf /tmp/novel_scraper_cache/*`
3. 检查网站是否可访问

### 章节号不正确
- 使用 V5 自动从标题解析章节号

### 文件保存位置
```bash
ls -lt ~/.openclaw/workspace/novels/ | head -5
```

---

**版本**: 1.4.0  
**更新**: 2026-04-02  
**详见**: CHANGELOG.md

Popular in Data Analysis

Data month: 2026-09 · Downloads, stars and installs are aggregated monthly from public skill registries (ClawHub, SkillHub). See methodology.