Skip to content

Releases: atomno-mcp/mcp-seo-audit

v0.4.3

Choose a tag to compare

@atomnos atomnos released this 06 Jul 12:40

Changed

  • Updated PyPI package metadata: author and contact email now use the atomno-mcp brand (hello@atomno-mcp.ru).

v0.4.2

Choose a tag to compare

@atomnos atomnos released this 05 Jul 17:11

Full Changelog: v0.4.1...v0.4.2

v0.4.1

Choose a tag to compare

@atomnos atomnos released this 04 Jul 05:23

chore: MCP Registry metadata (mcp-name marker in README)

v0.4.0 — audit_diff

Choose a tag to compare

@atomnos atomnos released this 23 Jun 16:45

Added

  • audit_diff(url, lang) — re-audits a site and compares it to the previous saved snapshot: health/score delta and which checks got worse or better. First call stores a baseline. Stateful PRO feature (requires DETAILWEB_API_KEY) — without a key returns available=false + upsell.

This is the first piece of the monitoring track and the concrete answer to "why not just ask an LLM?" — a language model cannot track a site's state over time.

Changed

  • Positioning «instrument, not a guess» in server instructions / audit_site description.
  • README: "Why not just ask an LLM?" + condensed Russian section.

Install: uvx atomno-mcp-seo-audit · PyPI: https://pypi.org/project/atomno-mcp-seo-audit/0.4.0/

v0.3.0 — first public release

Choose a tag to compare

@atomnos atomnos released this 23 Jun 14:13

First public release of atomno-mcp-seo-audit — a thin MCP (Model Context Protocol) server that gives AI agents technical SEO superpowers via the detail.web audit engine.

Tools

  • audit_site — full technical SEO audit: health score, letter grade, issues across 8 categories (HTTP/SSL, SEO meta, JSON-LD, Open Graph, sitemap/robots, images, performance, GEO).
  • list_checks — full free/PRO check catalogue (78 checks) grouped by category.
  • explain_issue — detailed why-it-matters and how-to-fix for a single check.
  • validate_robots — robots.txt validator (syntax, sitemap directive, blocked CSS/JS, AI-crawler blocks like GPTBot/ClaudeBot). Works on pasted content, no fetch.
  • check_sitemap — fetch + parse a sitemap (urlset vs index, URL count, common errors). SSRF-guarded.
  • build_jsonld — generate schema.org JSON-LD (Organization, LocalBusiness, Article, Product, FAQPage, BreadcrumbList, WebSite).
  • build_meta — generate <head> meta tags (title, description, canonical, Open Graph, Twitter Card) with length validation.

Notes

  • Bilingual output (ru|en) via lang parameter for titles/categories.
  • Free tier exposes the core check set; PRO unlocks the full suite (API key).
  • Backend: https://api.detailweb.ru. Install: pipx install atomno-mcp-seo-audit.

MIT licensed.