{ "markdown": "\n\n# atomno-mcp-seo-audit\n\n[![PyPI](https://img.shields.io/pypi/v/atomno-mcp-seo-audit)](https://pypi.org/project/atomno-mcp-seo-audit/)\n[![Python](https://img.shields.io/pypi/pyversions/atomno-mcp-seo-audit)](https://pypi.org/project/atomno-mcp-seo-audit/)\n[![License: MIT](https://img.shields.io/badge/license-MIT-green)](LICENSE)\n[![MCP](https://img.shields.io/badge/MCP-server-blue)](https://modelcontextprotocol.io)\n\nMCP (Model Context Protocol) server for **technical SEO & GEO audits**, powered by\nthe [detail.web](https://audit.detailweb.ru) engine — **real measurements, not LLM\nguesses**. Run it from Cursor, Claude Desktop or any MCP client. **8 tools:**\n`audit_site` (deterministic `0–100` health score + letter grade, 78 checks across\n8 categories, plus a **GEO** sub-score — visibility in ChatGPT / Perplexity /\nGoogle AI Overviews), `audit_diff` (compare vs the previous snapshot),\n`robots.txt` & `sitemap.xml` validators, JSON-LD & meta/OpenGraph builders, and\nper-check fix explainers. Probes TLS, redirects, TTFB, AI-crawler access (GPTBot)\nand `llms.txt`. Free tier + **PRO+** (deep-crawl, GEO, 40+ deeper checks) — API\nkeys unlock the deeper engine on the **PRO+** plan and up (see [Free vs PRO+](#free-vs-pro)).\n\n## Why pair it with an LLM?\n\nA language model on its own infers a site from training data and, at best, one\nrendered page — it can't directly read your TLS certificate's expiry, measure\nresponse time, parse `sitemap.xml`, or check whether `GPTBot` is blocked in\n`robots.txt`. This server runs those checks for real: actual HTTP requests,\nsecurity headers, redirect chains, structured-data validation — and returns a\n**deterministic** score (same site → same number), reproducible enough to put in\na client report. Think of it as the instrument and the LLM as the analyst that\ninterprets the readout — the two work best together.\n\n## What you get\n\n- **`audit_site(url, depth=1, lang=\"ru\")`** — one call returns:\n - health score `0–100` (higher is better) and a letter grade `A–F`;\n - issues grouped by category (security, SEO & indexing, performance, GEO, …),\n each with status `pass / warn / fail`;\n - a short human-readable summary.\n- **`audit_diff(url, lang=\"ru\")`** — re-audits a site and compares it to the\n previous run: health/score delta and which checks got worse or better. The\n first call stores a baseline. This is something a one-off LLM question can't\n do — track a site over time. Stateful feature — needs a key on the **PRO+**\n plan or higher.\n- **`list_checks(lang=\"ru\")`** — the full catalogue of engine checks grouped by\n category, with a `free` / `PRO` badge on each — so you (and the agent) can see\n exactly what the free tier covers and what PRO unlocks.\n- **`explain_issue(check_id, lang=\"ru\")`** — a deep-dive on a single check:\n why it matters and how to fix it. Pass a `check_id` from `audit_site` or\n `list_checks`. Title and category are localized; detailed advice is currently\n in Russian (the `advice_lang` field reports this).\n- **`validate_robots(content, lang=\"ru\")`** — paste a `robots.txt` and get back\n syntax issues, whether a `Sitemap:` directive is present, whether CSS/JS is\n blocked from render bots, and which AI crawlers (GPTBot, ClaudeBot, …) are\n explicitly blocked. No fetch — validates the text you provide.\n- **`check_sitemap(url, lang=\"ru\")`** — fetches a sitemap by URL and reports its\n format (`urlset` / `sitemapindex`), URL count and common problems (404,\n non-XML content type, `http://` links, missing ``, the 50k-per-file\n limit). The fetch is SSRF-guarded on the server.\n- **`build_jsonld(type, fields, lang=\"ru\")`** — generates a ready-to-paste\n schema.org JSON-LD `