Quality Checker

v1.0.2

章节质量检测器 - 检测章节正文的质量问题并生成评分报告。当需要评估章节质量、发现问题时使用,支持字数/段落/对话/标点/重复用词检测。

0· 107·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for yuzhihui886/quality-checker.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Quality Checker" (yuzhihui886/quality-checker) from ClawHub.
Skill page: https://clawhub.ai/yuzhihui886/quality-checker
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install quality-checker

ClawHub CLI

Package manager switcher

npx clawhub@latest install quality-checker
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description (chapter quality checker) match the included script and config. The files implement local text analysis (word counts, paragraph lengths, dialogue ratio, punctuation, repeated words). Declared dependencies (rich, PyYAML) and the single included Python script are proportionate to the stated functionality.
Instruction Scope
SKILL.md instructs running the included Python script against local Markdown/TXT files and saving reports locally. The runtime instructions reference only local files and the included config; there are no directives to read unrelated system files, environment variables, or to send data externally. Integration examples call another (presumably related) novel-writer script, which is consistent with a writing toolchain.
Install Mechanism
There is no install spec; skill is instruction-only but includes a script and a requirements.txt. Dependencies are common PyPI packages (rich, PyYAML) installed via pip—this is a standard, low-risk approach. No downloads from unknown URLs or archive extraction are present.
Credentials
The skill declares no required environment variables, credentials, or config paths. The code does not access environment variables or network endpoints; it only reads files whose paths are provided by the user and a local config file. Requested permissions are minimal and appropriate.
Persistence & Privilege
always is false (default), and the skill does not request persistent or elevated presence. It does not modify other skills or global agent settings. Behavior is limited to on-demand CLI execution.
Assessment
This skill appears to do exactly what it says: analyze local chapter files and generate a report. Before installing/run: - Only run the script on files you trust (it will read any filepath you pass). - Install dependencies in a virtual environment (pip install -r scripts/requirements.txt) to avoid contaminating your system Python. - Review configs/configs/quality_config.yml to adjust thresholds to your needs. - Note minor detection limitations (dialogue detection via quotes and Chinese word heuristics may produce false positives). Overall this is coherent and low-risk, but treat code from unknown sources cautiously (run in an isolated environment if you have concerns).

Like a lobster shell, security has layers — review code before you run it.

latestvk97cpjz2q41jjq6s0nb86cr7sx846x7q
107downloads
0stars
3versions
Updated 3w ago
v1.0.2
MIT-0

Quality Checker - 章节质量检测器

Overview

分析章节正文,检测潜在质量问题并生成详细报告,包含评分(0-100 分)和具体改进建议。

使用场景

  • 需要评估章节质量
  • 需要发现写作问题(字数不足、段落过长、对话比例异常等)
  • 需要量化质量指标进行对比
  • 需要在发布前进行质量把关

检测项目

检测项说明阈值
字数检测检查是否达到目标字数≥ 目标字数 80%
段落长度检测过长段落(影响阅读节奏)≤ 200 字/段
对话比例检测对话占比是否合理20%-60%
标点滥用检测感叹号/问号过度使用≤ 10%
重复用词检测高频重复词汇≤ 5 次/词

质量评分

分数等级说明
90-100优秀可直接发布
80-89良好少量问题
70-79中等需要修改
60-69及格建议修改
<60不合格需要大修

CLI 使用

# 基本用法
python3 scripts/check_quality.py --input chapters/chapter-01.md

# 指定目标字数
python3 scripts/check_quality.py \
  --input chapters/chapter-01.md \
  --word-count 3000

# 输出报告到文件
python3 scripts/check_quality.py \
  --input chapters/chapter-01.md \
  --output reports/chapter-01-quality.md

# 简写
python3 scripts/check_quality.py -i chapter.md -w 3000 -o report.md

参数说明

参数必填说明
--input / -i章节正文文件路径(Markdown)
--word-count / -w目标字数(默认 3000)
--output / -o输出报告文件路径

输出报告格式

# 章节质量检测报告

**章节**: chapter-01.md
**检测时间**: 2026-04-04 23:50:00
**质量评分**: 85/100 (良好)

## 检测结果

| 检测项 | 状态 | 详情 |
|--------|------|------|
| 字数 | ✅ 通过 | 2850/3000 (95%) |
| 段落长度 | ⚠️ 警告 | 3 个段落超过 200 字 |
| 对话比例 | ✅ 通过 | 35% (正常范围 20%-60%) |
| 标点滥用 | ✅ 通过 | 感叹号占比 3% |
| 重复用词 | ⚠️ 警告 | "突然" 出现 8 次 |

## 问题详情

### 段落过长
- 第 3 段:256 字
- 第 7 段:234 字
- 第 12 段:218 字

### 重复用词
- "突然": 8 次
- "仿佛": 6 次
- "感觉": 5 次

## 改进建议

1. 拆分第 3、7、12 段,每段控制在 200 字以内
2. 减少"突然"的使用频率,尝试同义替换
3. 增加 150 字达到目标字数

依赖

  • Python 3.8+
  • rich (终端渲染)

安装依赖:

pip install -r scripts/requirements.txt

与其他技能集成

与 novel-writer 集成

# 1. 生成章节正文
python3 ../novel-writer/scripts/write_chapter.py \
  --book-dir projects/my-novel \
  --chapter 1 \
  --outline outlines/chapter-1.md \
  --output chapters/chapter-01.md

# 2. 检测质量
python3 scripts/check_quality.py \
  --input chapters/chapter-01.md \
  --word-count 3000 \
  --output reports/chapter-01-quality.md

批量检测

# 检测全部章节
for chapter in chapters/chapter-*.md; do
    python3 scripts/check_quality.py \
        --input "$chapter" \
        --word-count 3000 \
        --output "reports/$(basename "$chapter" .md)-quality.md"
done

注意事项

  • 支持 UTF-8 和 GBK 编码的 Markdown 文件
  • 质量评分仅供参考,具体标准可根据项目调整
  • 对话检测基于引号识别,可能不完全准确
  • 重复用词检测自动过滤常见虚词(的、了、是等)

Comments

Loading comments...