內容簡介
SEOries Markdown Generator 是一款獨立的 WordPress 外掛，旨在為 AI 系統、大型語言模型（LLMs）和自動化 Markdown 消費者提供乾淨、結構化的 Markdown 表示。此外掛確保傳統訪客仍然獲得標準 HTML，同時保護 SEO 完整性。
【主要功能】
• Canonical URL 協商：透過標準 HTTP 協商提供 Markdown 內容
• AI Bot 自動交付：為識別的 AI 搜尋索引器自動提供 Markdown
• 搜尋引擎保護：主要搜尋引擎始終接收標準 HTML
• 存取控制與隱私：排除草稿、私密文章等內容
• 高效能快取：整合 WordPress 物件快取 API
• 管理員預覽與下載：可預覽生成的 Markdown 並下載
外掛標籤
開發者團隊
原文外掛簡介
SEOries Markdown Generator is an independent WordPress plugin developed by Mahmut Orkun Köksalan, built to provide clean, structured Markdown representations of WordPress content for AI systems, Large Language Models (LLMs), and automated markdown consumers.
Key Capabilities & Architectural Principles
Canonical URL Negotiation: Delivers text/markdown directly over canonical URLs via standard HTTP Accept: text/markdown content negotiation. Normal visitors and web browsers continue receiving standard HTML.
Public .md Policy: Direct public .md URLs (e.g. /post-slug.md) are disabled by default and return 404 Not Found to prevent URL fragmentation and preserve canonical SEO integrity.
AI Bot Automatic Delivery: Optional automated Markdown delivery for recognized AI search indexers and user-action bots (OAI-SearchBot, ChatGPT-User, Claude-SearchBot, Claude-User, PerplexityBot, Perplexity-User) and training bots (GPTBot, ClaudeBot).
Search Engine Protection: Major search engine crawlers (Googlebot, Bingbot, YandexBot) always receive standard HTML, ensuring zero interference with traditional search indexation.
Access Control & Privacy: Drafts, pending reviews, private posts, trashed items, and password-protected content are strictly excluded from public Markdown delivery.
Robots Directives & Noindex: Robots meta directives and headers are not access control mechanisms. Directives such as noindex do not prevent Markdown delivery; instead, robots directives are preserved and forwarded in the Markdown response headers (X-Robots-Tag) where applicable.
User-Agent Advisory: User-Agent matching is an ergonomic content-representation selector, not a security boundary or authentication mechanism.
DOM-Driven Content Extraction: Generic HTML parsing using native DOM & XPath, extracting clean semantic content without hard-coded theme or page-builder dependencies.
Safe CSS Selectors: Configure global, post-type, per-post (meta box), and archive-level main content and excluded selectors with built-in fail-safe protection.
High-Performance Caching: Integrates with WordPress Object Cache API (Redis / Memcached) with automatic Transients fallback, conditional HTTP (ETag / 304 Not Modified), and event-driven cache invalidation.
Buffer-Isolated Raw Delivery: Unified output buffer management ensuring newlines, frontmatter, and code fences are never corrupted by third-party HTML minifiers or theme output filters.
Admin Preview & Download: Authorized post editors can preview generated Markdown, inspect live extraction diagnostics, copy to clipboard, or download .md files directly from the WordPress editor.
Official Documentation: Comprehensive guides, bot user-agent tables, and integration tutorials are available at https://ai-markdown.seories.com/docs/
