Content Optimizer

io.github.sharozdawa由 GitHub 账号 sharozdawa 发布v1.0.0更新于 Oct 24, 2026

SERP-based content scoring and optimization with 7 SEO categories.

已验证STDIO仅桌面网页搜索与抓取数据与分析

概览

AI 生成的概览

一个本地 MCP 服务器，基于搜索结果对网页内容评分与优化，覆盖七个 SEO 类别并给出改进建议。

可以做什么
针对目标关键词对内容进行七个 SEO 类别的评分，给出 0-100 分及分类明细，并附按优先级排序的改进建议；同时提供可读性检查（Flesch-Kincaid、年级水平、句子统计、阅读时间）。分析搜索结果中排名靠前页面的字数、标题和主题基准，找出竞争对手覆盖而你的内容缺失的主题。检查标题结构并给出含关键词的标题改进建议。
适用场景
适合在撰写或修改 SEO 文章时使用，让助手把草稿与排名靠前的页面做对比、发现缺失主题，并获得具体的评分与可读性反馈，而无需付费订阅工具。
运行要求
通过 npm 包 content-optimizer-mcp 以 npx 在本地以 stdio 运行；需要 Node.js 18+ 和 npm 9+。未声明认证、环境变量或请求头。SERP 分析意味着需要网络访问以获取搜索结果。
安装前请注意
这些工具会分析你提供的内容，因此草稿和关键词会被发送到服务器查询 SERP 数据所用的搜索来源；在用于机密材料前请确认其访问的对象。未声明任何凭据或付费。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Content Optimizer，将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only，通过 STDIO。 STDIO 服务会启动本地进程，因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

[图片]

SERP-based content scoring, entity analysis, and actionable SEO recommendations.
Score your content against top-ranking pages. Know exactly what to improve.

[npm] [stars] [license] [PRs Welcome] [Content Optimizer MCP server]

Why Content Optimizer?

Paid content optimization tools charge $89-$170/month for SERP-based scoring. Content Optimizer gives you the same analysis — for free, self-hosted, and extensible.

FeatureContent OptimizerSurfer SEOClearscopeFrase
PriceFree$89/mo$170/mo$15/mo
Self-hosted:white_check_mark::x::x::x:
Open source:white_check_mark::x::x::x:
Content scoring:white_check_mark::white_check_mark::white_check_mark::white_check_mark:
Keyword analysis:white_check_mark::white_check_mark::white_check_mark::white_check_mark:
Readability analysis:white_check_mark::white_check_mark::white_check_mark::white_check_mark:
Heading optimization:white_check_mark::white_check_mark::white_check_mark::x:
Entity/topic coverage:white_check_mark::white_check_mark::white_check_mark::white_check_mark:
MCP server:white_check_mark::x::x::x:
AI editor integration:white_check_mark::x::x::x:
API access:white_check_mark:Paid add-on:x:Paid add-on
No data limits:white_check_mark:Credits-basedCredits-basedCredits-based

:sparkles: Features

:dart: Content Scoring — Score content 0-100 across 7 SEO categories with per-category breakdowns

:mag: SERP Analysis — Analyze top-ranking pages for word count, headings, and topic benchmarks

:abc: Readability Analysis — Flesch-Kincaid scoring, grade level, sentence length, reading time

:bookmark_tabs: Heading Optimization — Detect hierarchy issues, suggest keyword-rich heading structure

:brain: Entity Coverage — Find topics competitors cover that you're missing

:clipboard: Actionable Recommendations — Prioritized improvements ranked by impact

:robot: MCP Server — Use with Claude Desktop, Cursor, or any MCP-compatible AI tool

:chart_with_upwards_trend: Scoring Categories

CategoryMax PointsWhat It Measures
Keyword Usage20Density, placement in first paragraph and headings
Word Count15Content length vs SERP average and target
Heading Structure15H1/H2/H3 hierarchy, count, keyword inclusion
Readability15Flesch-Kincaid score, grade level
Entity Coverage15Topic coverage vs SERP competitors
Content Depth10Paragraphs, lists, statistics, data points
Internal Structure10Introduction, conclusion, logical flow
100

:camera: Screenshots

Screenshots coming soon — the web UI is under active development.

:rocket: Self-Hosting Setup

Prerequisites

  • Node.js 18+
  • npm 9+

Installation

bash
git clone https://github.com/sharozdawa/content-optimizer.gitcd content-optimizernpm install

Development

bash
npm run dev

Open http://localhost:3000.

Production Build

bash
npm run buildnpm start

:electric_plug: MCP Server

The MCP server lets you use Content Optimizer directly inside AI coding tools like Claude Desktop and Cursor.

Install via npx

No setup required — just add the config:

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

json
{  "mcpServers": {    "content-optimizer": {      "command": "npx",      "args": ["-y", "content-optimizer-mcp"]    }  }}

Cursor

Add to .cursor/mcp.json in your project:

json
{  "mcpServers": {    "content-optimizer": {      "command": "npx",      "args": ["-y", "content-optimizer-mcp"]    }  }}

Build from Source

bash
cd mcp-servernpm installnpm run build

Then use node ./mcp-server/dist/index.js as the command in your MCP config.

MCP Tools

ToolDescription
score_contentScore content against a keyword across 7 categories (0-100)
analyze_keywordAnalyze SERP data — top 10 results with word counts, headings, topics
get_content_recommendationsGet prioritized improvement recommendations
check_readabilityFlesch-Kincaid score, grade level, sentence stats, reading time
find_missing_topicsFind topics from SERP competitors missing in your content
optimize_headingsAnalyze heading structure and suggest improvements

:open_book: More Open Source SEO Tools

ProjectDescription
Schema GeneratorGenerate structured data (JSON-LD) for any page — Product, Article, FAQ, HowTo, and more
AI VisibilityTrack how your brand appears in AI search engines (ChatGPT, Perplexity, Gemini)
IndexNow MCPInstantly notify search engines when you publish or update content
HumanPenAI content humanizer — make AI-generated text sound natural

:handshake: Contributing

Contributions are welcome! See CONTRIBUTING.md for guidelines.

:page_facing_up: License

MIT — see LICENSE for details.

Built by Sharoz Dawa
[GitHub]

[图片]

来源：README.md，提交 96a551a

工具

0
SourceWeft 尚未在沙箱中启动过该软件包，因此其工具未知。

版本历史

1
  1. v1.0.0最新Sep 16, 2026

评分

已满足适用检查项的 50%

该评分衡量此收录条目的完整程度，依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价，也不表示服务器安全或当前可用：可连通性是带日期的检查结果，评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分

    请保持端点能够响应 MCP 握手；SourceWeft 每 30 天探测一次。

  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头，或保持端点能够响应探测，以便判定其认证方式。

  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list，或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分