mcp-seo-audit

io.github.atomno-mcp由 GitHub 账号 atomno-mcp 发布v0.4.7更新于 Oct 24, 2026

Technical SEO site audit MCP: health score, issues, robots, sitemap, JSON-LD.

发布方自己的服务:audit.domains - Domain Valuation

概览

AI 生成的概览

对网站执行真实的技术 SEO 与 GEO 审计,返回确定性的健康评分、分类问题与修复建议。

可以做什么
提供八个工具:audit_site 返回 0-100 健康评分、字母等级和按类别分组的问题;audit_diff 与上一次快照对比;list_checks 和 explain_issue 说明引擎的检查项及修复方法;validate_robots 和 check_sitemap 检查 robots.txt 与 sitemap.xml;build_jsonld 和 build_meta 生成 schema.org JSON-LD 与 head 元标签。它发起真实 HTTP 请求,探测 TLS、重定向、TTFB、AI 爬虫访问和 llms.txt,而非依赖模型知识推断。
适用场景
适合需要可测量、可复现的 SEO 结论而非猜测的场景,也适合跟踪两次审计之间的变化,或生成并校验结构化数据与元标签。
运行要求
通过 stdio 在本地运行,通常用 uvx 启动 PyPI 包 atomno-mcp-seo-audit,因此需要 Python/uv 运行时和网络访问。免费层无需密钥;更深入的检查、GEO 子评分、深度抓取和 audit_diff 需要来自 PRO+ 或更高套餐的 DETAILWEB_API_KEY。可选变量:DETAILWEB_API_BASE、DETAILWEB_TIMEOUT、DETAILWEB_LANG。
安装前请注意
可选的 DETAILWEB_API_KEY 是机密,应通过环境变量提供,不要外泄。审计会把目标 URL 发送到 detail.web 后端,站点地图检查会抓取远程 URL。免费层仅覆盖核心检查与单页;入门级 PRO 套餐的密钥仍只返回免费结果。详细修复建议目前为俄语。

安装

在 SourceWeft 中

  1. 打开 控制台中的 mcp-seo-audit,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

atomno-mcp-seo-audit

[PyPI] [Python] [License: MIT] [MCP]

MCP (Model Context Protocol) server for technical SEO & GEO audits, powered by the detail.web engine — real measurements, not LLM guesses. Run it from Cursor, Claude Desktop or any MCP client. 8 tools: audit_site (deterministic 0–100 health score + letter grade, 78 checks across 8 categories, plus a GEO sub-score — visibility in ChatGPT / Perplexity / Google AI Overviews), audit_diff (compare vs the previous snapshot), robots.txt & sitemap.xml validators, JSON-LD & meta/OpenGraph builders, and per-check fix explainers. Probes TLS, redirects, TTFB, AI-crawler access (GPTBot) and llms.txt. Free tier + PRO+ (deep-crawl, GEO, 40+ deeper checks) — API keys unlock the deeper engine on the PRO+ plan and up (see Free vs PRO+).

Why pair it with an LLM?

A language model on its own infers a site from training data and, at best, one rendered page — it can't directly read your TLS certificate's expiry, measure response time, parse sitemap.xml, or check whether GPTBot is blocked in robots.txt. This server runs those checks for real: actual HTTP requests, security headers, redirect chains, structured-data validation — and returns a deterministic score (same site → same number), reproducible enough to put in a client report. Think of it as the instrument and the LLM as the analyst that interprets the readout — the two work best together.

What you get

  • audit_site(url, depth=1, lang="ru") — one call returns:
    • health score 0–100 (higher is better) and a letter grade A–F;
    • issues grouped by category (security, SEO & indexing, performance, GEO, …), each with status pass / warn / fail;
    • a short human-readable summary.
  • audit_diff(url, lang="ru") — re-audits a site and compares it to the previous run: health/score delta and which checks got worse or better. The first call stores a baseline. This is something a one-off LLM question can't do — track a site over time. Stateful feature — needs a key on the PRO+ plan or higher.
  • list_checks(lang="ru") — the full catalogue of engine checks grouped by category, with a free / PRO badge on each — so you (and the agent) can see exactly what the free tier covers and what PRO unlocks.
  • explain_issue(check_id, lang="ru") — a deep-dive on a single check: why it matters and how to fix it. Pass a check_id from audit_site or list_checks. Title and category are localized; detailed advice is currently in Russian (the advice_lang field reports this).
  • validate_robots(content, lang="ru") — paste a robots.txt and get back syntax issues, whether a Sitemap: directive is present, whether CSS/JS is blocked from render bots, and which AI crawlers (GPTBot, ClaudeBot, …) are explicitly blocked. No fetch — validates the text you provide.
  • check_sitemap(url, lang="ru") — fetches a sitemap by URL and reports its format (urlset / sitemapindex), URL count and common problems (404, non-XML content type, http:// links, missing <lastmod>, the 50k-per-file limit). The fetch is SSRF-guarded on the server.
  • build_jsonld(type, fields, lang="ru") — generates a ready-to-paste schema.org JSON-LD <script> (Organization, LocalBusiness, Article, Product, FAQPage, BreadcrumbList, WebSite) and tells you which required/recommended fields are missing. It never invents data — only what you pass in.
  • build_meta(fields, lang="ru") — generates <head> meta tags (title, description, canonical, Open Graph, Twitter Card) and validates the title (50–60 chars) and description (120–160 chars) lengths.

Free vs PRO+

Free (no key)PRO+ and up (API key)
Checkscore technical basics40+ deeper checks (E-E-A-T, Schema.org, Goldmine title)
GEO4 GEO signalsGEO readiness sub-score + deep GEO checks
Crawlsingle pagedeep-crawl up to 20 pages (depth=2/3)

Which plan unlocks the API/MCP? Programmatic access (this server, audit_diff, deep-crawl, GEO sub-score) is enabled on PRO+ (pro_plus), Business and Enterprise. The entry-level PRO plan and the Free tier are web-dashboard only — an API key issued on them authenticates but still returns the free result. If you need programmatic access, pick PRO+ or higher.

The audit engine itself stays on the server — this package is a thin client (HTTP calls + formatting only).

Install

bash
uvx atomno-mcp-seo-audit

Or add to your MCP client config (mcp.json):

json
{  "mcpServers": {    "seo-audit": {      "command": "uvx",      "args": ["atomno-mcp-seo-audit"]    }  }}

Configuration

All via environment variables:

VariableDefaultPurpose
DETAILWEB_API_BASEhttps://api.detailweb.ruBackend base URL
DETAILWEB_API_KEY—API key (dwa_...) from a PRO+ plan or higher. Without it (or on Free/PRO) → free tier
DETAILWEB_TIMEOUT60HTTP timeout (seconds)
DETAILWEB_LANGruDefault issue-title language (ru / en)

The free tier needs no key and no signup — just run the command above. Programmatic access (40+ deeper checks, GEO sub-score, deep-crawl, audit_diff) requires a key from the PRO+ plan or higher — the entry-level PRO plan is web-dashboard only and its key returns the free result. It is currently provisioned on request: email [email protected] or reach out via audit.detailweb.ru. Once your account is active you create keys yourself in Dashboard → Account → API keys (dwa_…, shown once) and put the key in DETAILWEB_API_KEY.

Example

"Audit https://example.com"

The agent calls audit_site("https://example.com") and gets back the health score, grade and the list of issues to fix.

License

MIT © atomno-mcp. The open-source client talks to a proprietary hosted backend.


🇷🇺 На русском

MCP-сервер технического SEO-аудита на движке detail.web. Запускайте аудит прямо из ИИ-агента (Cursor, Claude Desktop и любой MCP-клиент): health-score, проблемы по 8 категориям и GEO-суб-балл (видимость в ИИ-поиске — ChatGPT, Perplexity, AI Overviews).

Зачем в связке с нейросетью. Языковая модель сама по себе судит о сайте по обучающим данным и в лучшем случае по одной отрисованной странице — она не прочитает напрямую срок SSL-сертификата, не измерит время ответа, не распарсит sitemap.xml и не проверит, заблокирован ли GPTBot в robots.txt. Этот сервер выполняет такие проверки по-настоящему: HTTP-запросы, заголовки, редиректы, микроразметка — и даёт детерминированный score (тот же сайт → то же число), пригодный для отчёта клиенту. Это прибор, а нейросеть — аналитик, который читает показания. Лучше всего работает связка.

Инструменты: audit_site (аудит + score + GEO), audit_diff (что изменилось с прошлой проверки — stateful, тариф PRO+ и выше), list_checks (каталог проверок free/PRO), explain_issue (почему важно + как исправить), validate_robots, check_sitemap, build_jsonld, build_meta.

Установка:

bash
uvx atomno-mcp-seo-audit

В конфиге MCP-клиента (mcp.json):

json
{  "mcpServers": {    "seo-audit": {      "command": "uvx",      "args": ["atomno-mcp-seo-audit"],      "env": { "DETAILWEB_LANG": "ru" }    }  }}

Бесплатный тариф (базовые проверки, одна страница) работает сразу, без ключа и регистрации. Программный доступ (40+ глубоких проверок, GEO-суб-балл, deep-crawl до 20 страниц, audit_diff) работает с ключом тарифа PRO+ (pro_plus) и выше — Business, Enterprise. Начальный тариф PRO (1290 ₽) и Free — только веб-кабинет: ключ на них проходит авторизацию, но результат остаётся бесплатным. Тариф пока выдаём по запросу: напишите на [email protected] или через audit.detailweb.ru. После активации аккаунта ключ (dwa_…) создаётся в кабинете → Аккаунт → API-ключи (показывается один раз) и подставляется в DETAILWEB_API_KEY в env. Полное описание инструментов и настроек — в английской версии выше.

来源:README.md,提交 751e636

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. v0.4.7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

其他 Audit 集成