Content Audit Expert
Overview
Content Audit Expert is a comprehensive AI-powered analysis tool that evaluates your written content across multiple dimensions—readability, tone, sentiment, SEO performance, and audience engagement potential. Rather than surface-level feedback, this skill provides deep structural analysis and actionable repurposing strategies.
This skill is invaluable for content teams, marketers, copywriters, and publishers who need to:
- - Maintain consistent brand voice across multiple channels (WordPress, Medium, LinkedIn, Slack)
- Optimize content for search engines with keyword analysis and heading structure recommendations
- Identify underperforming assets in your content library and discover repurposing opportunities
- Improve readability scores for diverse audience segments
- Detect tone misalignment before publishing across corporate, marketing, or support channels
- Analyze sentiment patterns to understand emotional resonance with your target audience
Unlike basic grammar checkers, Content Audit Expert leverages advanced NLP analysis to evaluate content holistically—examining narrative structure, keyword density, audience alignment, and conversion potential. It integrates seamlessly with WordPress editorial workflows, Slack publishing pipelines, and Google Drive content repositories.
Quick Start
Try these prompts immediately to see the skill in action:
Example 1: Blog Post Audit
CODEBLOCK0
Example 2: Email Copy Analysis
CODEBLOCK1
Example 3: Bulk Content Audit
CODEBLOCK2
Capabilities
1. Readability Analysis
- - Flesch Kincaid Grade Level calculation
- Gunning Fog Index evaluation
- Sentence length variance detection
- Paragraph structure optimization
- Passive vs. active voice ratio analysis
- Jargon and complexity flagging
Usage Example:
CODEBLOCK3
2. Tone & Voice Detection
- - Brand voice consistency scoring (0-100)
- Formal vs. conversational balance measurement
- Authority vs. approachability alignment
- Professional tone assessment
- Personality/character consistency tracking across multiple pieces
- Emotional resonance indicators
Usage Example:
CODEBLOCK4
3. Sentiment & Emotional Analysis
- - Overall sentiment classification (positive, negative, neutral)
- Emotion detection (joy, trust, fear, surprise, sadness, disgust, anger, anticipation)
- Emotional intensity scoring
- Sentiment progression tracking through content
- Reader emotional journey mapping
Usage Example:
CODEBLOCK5
4. SEO & Keyword Performance
- - Primary and secondary keyword identification
- Keyword density analysis with optimization recommendations
- Heading structure (H1/H2/H3) evaluation
- Meta description effectiveness
- Internal linking opportunity detection
- Target audience keyword intent alignment
Usage Example:
CODEBLOCK6
5. Content Repurposing Strategy
- - Identify evergreen vs. time-sensitive content
- Generate 5+ repurposing ideas (social posts, email sequences, videos, infographics, podcasts)
- Calculate ROI potential for each repurposing avenue
- Suggest format conversions (blog → email series, article → social carousel)
- Cross-promotion opportunity mapping
- Content fragmentation strategy (break long-form into series)
Usage Example:
CODEBLOCK7
6. Audience Alignment Scoring
- - Target persona matching (0-100)
- Language difficulty vs. audience reading level
- Relevance score to stated audience
- Engagement prediction based on demographic research
- Value prop clarity for specific personas
7. Competitive Benchmarking
- - Compare against industry standard readability scores
- Identify unique value propositions vs. competitor content
- Content depth analysis (word count, comprehensiveness)
- Missing information gaps relative to top-ranking competitors
Configuration
Required Environment Variables
CODEBLOCK8
Setup Instructions
- 1. Install dependencies (if running locally):
CODEBLOCK9
- 2. Authenticate with OpenAI:
- Get your API key from https://platform.openai.com/api-keys
- Set the
OPENAI_API_KEY environment variable
- Verify access: INLINECODE1
- 3. Integration Setups:
-
WordPress: Use the ClawHub WordPress plugin to send content directly from editor
-
Google Docs: Share document, provide link in prompt
-
Slack: Enable ClawHub in your Slack workspace, use
/audit [content] command
-
Email: Forward content via email or paste directly
Configuration Options
| Option | Default | Description |
|---|
| INLINECODE3 | comprehensive | basic, standard, or comprehensive analysis |
| INLINECODE4 |
general | Specify audience (C-suite, marketers, developers, etc.) |
|
CONTENT_FORMAT | article | article, email, social, whitepaper, case-study |
|
BRAND_VOICE_GUIDELINES | none | Paste your brand voice document for consistency scoring |
|
COMPETITOR_URLS | none | Include URLs for competitive benchmarking |
Example Outputs
Sample Readability Report
CODEBLOCK10
Sample Tone & Sentiment Report
CODEBLOCK11
Sample Repurposing Strategy
CODEBLOCK12
Tips & Best Practices
1. Establish Baseline Metrics
Before making changes, run an audit to establish your content quality baseline. Track these metrics over time:
- - Average readability score
- Sentiment consistency
- Brand voice alignment percentage
- SEO keyword density
This creates accountability and shows improvement ROI to stakeholders.
2. Use Comparative Audits
CODEBLOCK13
3. Implement Brand Voice Guidelines
Create a 1-page brand voice document covering:
- - Tone descriptors (warm, authoritative, playful, etc.)
- Vocabulary (preferred terms vs. forbidden jargon)
- Sentence structure preference
- Common phrases/mantras
- Examples of on-brand and off-brand writing
Run audits with this document for 95%+ consistency scoring.
4. Audit Before Updating Content
Always audit existing content before revisions. This:
- - Identifies what's currently working
- Prevents inadvertent tone shifts
- Shows specific improvement targets
- Creates a before/after comparison for stakeholder reporting
5. Leverage Repurposing for Content Multiplication
A single 2,000-word blog post can generate:
- - 1 email sequence (5-7 emails)
- 8-10 social media posts
- 1 webinar outline
- 1 short video script
- 1 podcast talking points outline
Multiplies content ROI by 5-8x without starting from scratch.
6. Segment Audits by Content Type
Different content types have different standards:
- - Blog posts: 70+ readability, positive sentiment
- Sales pages: 60-70 readability (exclusivity), high trust/anticipation
- Support docs: 75+ readability (clarity), neutral tone
- Email: 65-75 readability (scannable), warm tone
Configure audits for specific content type standards.
7. Monitor Sentiment Drift
If publishing frequently to a channel (newsletter, blog, social), track sentiment patterns:
- - Avoid 3+ consecutive negative posts
- Mix positive, neutral, and aspirational content
- Balance problem-focused with solution-focused messaging
- Maintain 70%+ positive sentiment overall for brand reputation
8. Use Competitive Benchmarking
Provide competitor URLs for comparative analysis:
CODEBLOCK14
Safety & Guardrails
What This Skill WILL Do
✓ Analyze readability, tone, sentiment, and keyword performance
✓ Suggest structural improvements and repurposing strategies
✓ Provide competitive benchmarking and gap analysis
✓ Generate specific, actionable recommendations
✓ Create before/after editing examples
What This Skill WILL NOT Do
✗
Not a substitute for subject matter expertise — The skill analyzes structure and style, not factual accuracy. Always verify claims, statistics, and technical details independently.
✗ Not a plagiarism detector — Use Copyscape, Turnitin, or Grammarly's plagiarism checker instead.
✗ Not for automated publishing — All recommendations require human review before publishing. The skill flags concerns but doesn't make editorial decisions.
✗ Not a replacement for editors — Use for optimization, not as a complete editing solution. Complex rewrites and narrative restructuring still benefit from human editors.
✗ Not legal or medical content validation — This skill cannot verify compliance with industry regulations (HIPAA, legal disclaimers, medical claims). Consult domain experts for regulated content.
✗ Not cultural sensitivity analysis — While the skill detects tone, it may miss cultural context or offensive implications. Review with subject matter experts familiar with your audience's culture.
Limitations & Boundaries
| Limitation | Workaround |
|---|
| Works best with 200+ words | Use bulk audits for shorter content; batch multiple pieces |
| Doesn't verify citations/facts |
Cross-reference claims with fact-checking tools independently |
| Brand voice matching requires clear guidelines | Provide detailed brand voice document for accuracy
技能名称:利用Google Sheets分析与Slack报告自动化内容审核
详细描述:
内容审核专家
概述
内容审核专家是一款全面的AI驱动分析工具,可从多个维度评估您的书面内容——可读性、语气、情感、SEO表现和受众参与潜力。该技能不仅提供表面反馈,还能进行深度结构分析并提供可操作的再利用策略。
此技能对于需要以下内容的内容团队、营销人员、文案撰写人员和出版商来说极具价值:
- - 在多个渠道(WordPress、Medium、LinkedIn、Slack)保持一致品牌声音
- 通过关键词分析和标题结构建议优化搜索引擎内容
- 识别内容库中表现不佳的资产并发现再利用机会
- 为不同受众群体提高可读性评分
- 在企业、营销或支持渠道发布前检测语气偏差
- 分析情感模式以了解与目标受众的情感共鸣
与基础语法检查器不同,内容审核专家利用先进的NLP分析全面评估内容——审视叙事结构、关键词密度、受众契合度和转化潜力。它与WordPress编辑工作流、Slack发布管道和Google Drive内容存储库无缝集成。
快速入门
立即尝试以下提示以体验该技能的实际应用:
示例1:博客文章审核
审核这篇博客文章的内容质量:
标题:提升生产力的10种方法
生产力很重要。每个人都想变得更高效。在本文中,
我们将讨论10种提升生产力的方法。这些方法经过验证
且有效。让我们开始吧。首先,你应该早起。早起能帮助
你完成更多事情。其次,你应该制定待办事项清单。清单能帮你保持条理。
第三,你应该消除干扰。干扰会阻碍你工作。
我们推荐使用Focus@Will或Forest等应用。
提供可读性评分、语气分析、情感细分以及3个具体的再利用建议。
示例2:邮件文案分析
分析这封邮件以优化转化率:
主题:查看我们的新产品
正文:你好,我们推出了一款新产品。它非常好。你应该看看。
点击此处了解更多。谢谢。
包括:情感分析、受众契合度评分、CTA有效性评级,
以及具体的文案改进建议(附前后对比示例)。
示例3:批量内容审核
对以下三篇文章进行内容审核,并识别交叉推广机会:
文章1(2400字):2024年SEO完全指南
文章2(800字):5个常见SEO错误
文章3(1200字):如何优化标题标签
提供:重叠分析、每篇文章的独特价值主张、再利用建议
(社交片段、邮件序列、LinkedIn帖子)以及内容缺口报告。
能力
1. 可读性分析
- - Flesch Kincaid年级水平计算
- Gunning Fog指数评估
- 句子长度变化检测
- 段落结构优化
- 被动语态与主动语态比例分析
- 行话和复杂性标记
使用示例:
评估面向C级高管的 technical 白皮书的可读性,
目标年级水平为10-12级。标记任何平均每句超过18个单词的过于复杂的段落。
2. 语气与声音检测
- - 品牌声音一致性评分(0-100)
- 正式与对话式平衡测量
- 权威性与亲和力契合度
- 专业语气评估
- 跨多篇文章的个性/角色一致性追踪
- 情感共鸣指标
使用示例:
分析这封客户支持邮件是否保持我们的品牌声音。
我们的品牌声音指南:温暖、专业、以解决方案为导向、富有同理心。
标记任何偏离这些原则的地方。
3. 情感与情绪分析
- - 整体情感分类(正面、负面、中性)
- 情绪检测(喜悦、信任、恐惧、惊讶、悲伤、厌恶、愤怒、期待)
- 情绪强度评分
- 内容中的情感进展追踪
- 读者情感旅程映射
使用示例:
情感审核:这个产品落地页是否唤起了正确的情绪
(自信、兴奋、信任)以推动转化?逐段映射情感旅程。
4. SEO与关键词表现
- - 主要和次要关键词识别
- 关键词密度分析与优化建议
- 标题结构(H1/H2/H3)评估
- 元描述有效性
- 内部链接机会检测
- 目标受众关键词意图契合度
使用示例:
针对目标关键词进行SEO审核:内容营销、营销自动化、
内容日历。评估关键词位置、密度,并建议H2/H3
结构调整以获得更好的SERP可见性。
5. 内容再利用策略
- - 识别常青内容与时效性内容
- 生成5个以上的再利用建议(社交帖子、邮件序列、视频、信息图、播客)
- 计算每个再利用途径的ROI潜力
- 建议格式转换(博客→邮件系列、文章→社交轮播)
- 交叉推广机会映射
- 内容碎片化策略(将长文拆分为系列)
使用示例:
提出8种方式来再利用这篇2000字的指南。包括:LinkedIn文章
片段、Twitter/X线程大纲、邮件序列结构、YouTube视频
大纲、播客讨论要点、信息图数据点和网络研讨会文字记录。
6. 受众契合度评分
- - 目标人物匹配度(0-100)
- 语言难度与受众阅读水平对比
- 与所述受众的相关性评分
- 基于人口统计研究的参与度预测
- 针对特定人物的价值主张清晰度
7. 竞争基准测试
- - 与行业标准可读性评分对比
- 识别与竞争对手内容相比的独特价值主张
- 内容深度分析(字数、全面性)
- 相对于排名靠前的竞争对手的信息缺失缺口
配置
所需环境变量
bash
设置您的OpenAI API密钥(AI分析必需)
export OPENAI
APIKEY=sk-...
可选:设置分析的首选温度(建议0.3-0.7)
export AUDIT_TEMPERATURE=0.5
可选:设置内容分析深度(基础、标准、全面)
export AUDIT_DEPTH=comprehensive
设置说明
- 1. 安装依赖项(如果在本地运行):
bash
pip install openai python-dotenv textstat
- 2. 向OpenAI进行身份验证:
- 从https://platform.openai.com/api-keys获取您的API密钥
- 设置OPENAI
APIKEY环境变量
- 验证访问权限:curl https://api.openai.com/v1/models -H Authorization: Bearer $OPENAI
APIKEY
- 3. 集成设置:
-
WordPress:使用ClawHub WordPress插件直接从编辑器发送内容
-
Google Docs:分享文档,在提示中提供链接
-
Slack:在您的Slack工作区启用ClawHub,使用/audit [内容]命令
-
电子邮件:通过电子邮件转发内容或直接粘贴
配置选项
| 选项 | 默认值 | 描述 |
|---|
| AUDITDEPTH | comprehensive | 基础、标准或全面分析 |
| TARGETAUDIENCE |
general | 指定受众(C级高管、营销人员、开发者等) |
| CONTENT_FORMAT | article | 文章、邮件、社交、白皮书、案例研究 |
| BRAND
VOICEGUIDELINES | none | 粘贴您的品牌声音文档以进行一致性评分 |
| COMPETITOR_URLS | none | 包含用于竞争基准测试的URL |
示例输出
样本可读性报告
═══════════════════════════════════════════════════════════════
可读性分析
═══════════════════════════════════════════════════════════════
整体可读性评分:72/100(良好)
目标年级水平:9.4(高中二年级)
Flesch Kincaid年级:10.2
细分:
├─ 句子多样性:85/100 ✓(范围:8-32个单词)
├─ 段落长度:68/100 ⚠(平均:4.2个句子)
├─ 被动语态:15%(目标:<20%)✓
├─ 行话密度:8%(可接受)
└─ 单词复杂度:92%常用词 ✓
首要建议:
- 1. 将第3段拆分为2个较短的段落(目前6个句子)
- 将第47行的leverage简化为use
- 为第5段添加副标题以增强结构
样本语气与情感报告
═══════════════════════════════════════════════════════════════
语气与情感分析
═══════════════════════════════════════════════════════════════
整体情感:正面(78%置信度)
语气分类:专业 + 对话式(平衡良好)
情绪细分:
├─ 信任:82%(强烈的权威信号)
├─ 期待:71%(前瞻性语言)
├─ 喜悦:56%(中等积极性)
├─ 恐惧:12