Ambiscale

<p>As AI tools like ChatGPT, Claude, and Perplexity change how people search, <strong>Ambiscale LLMs Friendly</strong> plugin ensures your website is accurately represented – enhancing visibility and accuracy in AI-driven recommendations.</p>
<h3>Why do I need it?</h3>
<p>With AI tools playing a growing role in content discovery and analysis, websites need a reliable, standardized way to communicate with large language models. Our plugin takes care of this automatically – generating structured, LLM-friendly text files that help AI systems accurately interpret and present your WordPress content.</p>
<h3>What is llms.txt</h3>
<p>The llms.txt (similar to robots.txt) is a proposed standard for helping large language models understand website content during inference time. It provides a structured markdown file called llms.txt containing key information about your site, making it easier for AI tools to:</p>
<ul>
<li>Quickly understand what your site is about</li>
<li>Access the most important content</li>
<li>Navigate your site’s structure efficiently</li>
<li>Work within context window limitations</li>
</ul>
<h3>When to use LLMs.txt</h3>
<ul>
<li><strong>Developer documentation</strong> – Ensure AI coding assistants can easily access and understand your technical content.  </li>
<li><strong>Business websites</strong> – Help AI tools accurately interpret your products, services, and company details.  </li>
<li><strong>Educational resources</strong> – Make it easier for AI tutors to explain and navigate your learning materials.  </li>
<li><strong>News & blogs</strong> – Allow AI to summarize, reference, and contextualize your articles with greater precision.  </li>
<li><strong>E-commerce sites</strong> – Enable AI shopping assistants to read and reason over your product catalog.  </li>
<li><strong>Personal portfolios</strong> – Showcase your work and achievements in a format that’s accessible to AI models.</li>
</ul>
<h3>Key features</h3>
<h3>Automatic llms.txt Generation</h3>
<ul>
<li>Creates <code>/llms.txt</code> file following the official specification format</li>
<li>Includes site name, summary, and detailed description</li>
<li>Organizes content by post types with customizable headings</li>
<li>Automatically updates when content changes</li>
</ul>
<h3>Extended Content Support</h3>
<ul>
<li>Generates optional <code>/llms-full.txt</code> with complete post content</li>
<li>Creates individual markdown files for each post</li>
<li>Supports selective export of specific posts</li>
<li>Maintains clean, AI-readable markdown format</li>
</ul>
<h3>Markdown Page Endpoints</h3>
<ul>
<li>Adds <code>?format=md</code> parameter to any post/page URL for markdown version</li>
<li>Implements the llms.txt specification’s <code>.md</code> URL pattern proposal</li>
<li>Blocks search engine bots from accessing markdown versions</li>
<li>Perfect for development environments and API documentation</li>
</ul>
<h3>Flexible Configuration</h3>
<ul>
<li>Choose which post types to include (posts, pages, custom post types)</li>
<li>Customize section headings for each post type</li>
<li>Set custom excerpts specifically for LLM consumption</li>
<li>Control excerpt generation for posts without manual excerpts</li>
</ul>
<h3>Advanced File Management</h3>
<ul>
<li>Choose export location: uploads, wp-content, or root directory</li>
<li>UTF-8 encoding with BOM for maximum compatibility</li>
<li>Automatic cleanup when changing export directories</li>
<li>WordPress Filesystem API integration for security</li>
</ul>
<h3>SEO Integration</h3>
<ul>
<li>Full Yoast SEO sitemap integration</li>
<li>Adds llms.txt files to XML sitemap</li>
<li>Proper lastmod timestamps and priorities</li>
<li>Helps search engines discover your LLM-friendly content</li>
</ul>
<h3>Enterprise Security</h3>
<ul>
<li>Comprehensive security headers implementation</li>
<li>Rate limiting for file generation</li>
<li>Security event logging with IP tracking</li>
<li>Capability-based access controls</li>
<li>Input sanitization and output escaping</li>
</ul>
<h3>Modern Admin Interface</h3>
<ul>
<li>Clean, intuitive settings page with tabs</li>
<li>Live preview of generated files</li>
<li>File status monitoring</li>
<li>One-click regeneration</li>
<li>Responsive design</li>
</ul>
<p>Company site: <a href="https://ambiscale.com" rel="nofollow ugc">Ambiscale</a></p>
<h3>Privacy Policy</h3>
<p>This plugin does not:<br />
– Collect any personal data<br />
– Send data to external servers<br />
– Use cookies<br />
– Track users</p>
<p>All generated files contain only publicly available content from your WordPress site.</p>
<h3>Credits</h3>
<ul>
<li>Developed by <a href="https://ambiscale.com" rel="nofollow ugc">Ambiscale</a></li>
<li>Uses League\HTMLToMarkdown for markdown conversion</li>
<li>Icons and graphics by Ambiscale design team</li>
</ul>


Ambiscale LLMs Friendly

Implements the llms.txt specification to make your WordPress content AI-friendly and optimized for Large Language Models (LLMs).

phalkmin

<p>ParseLess serves your WordPress content as clean Markdown to AI crawlers (via User-Agent detection) and on manual <code>?format=md</code> requests. Same URL, same content, none of the theme chrome, navigation, widgets, or page-builder scaffolding that AI bots and CLI tools don’t need.</p>
<p>It also exposes <code>/llms.txt</code> for the emerging AI-indexing standard, so models know where to find your content.</p>
<h4>Who is this for?</h4>
<p><strong>1. Developers using Claude Code, Cursor, Aider, or any CLI that feeds your own site content into an LLM.</strong></p>
<p>If you’ve ever piped a blog post into Claude or ChatGPT for analysis, rewriting, or summarization, you’ve watched it burn through tokens parsing nav menus, footer markup, and CSS classes that have nothing to do with your content. A typical WordPress page measured live: <strong>~19,800 tokens of HTML for ~975 tokens of actual content</strong> — a 20x reduction just by stripping the theme. Heavy page-builder sites (Elementor, Divi) routinely hit 100x or more.</p>
<p>That’s the difference between fitting a handful of pages in a context window and fitting dozens.</p>
<p>Just append <code>?format=md</code> to any post URL and pipe it straight into your tool of choice:</p>
<pre><code>curl https://yoursite.com/my-post/?format=md | claude "summarize this"
</code></pre>
<p><strong>2. Site owners getting “high resource usage” warnings from their host because AI bots are hammering the site.</strong></p>
<p>GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and a dozen others crawl WordPress sites constantly. Each request renders your full theme, runs widget queries, loads page-builder assets, and ships hundreds of KB of HTML per page — most of which the bot discards before extracting the actual text.</p>
<p>ParseLess intercepts these crawls and serves a tiny Markdown payload instead. You keep the AEO/SEO benefit of being indexed by AI search, without paying the server cost of rendering your full theme for every bot hit.</p>
<p>Typical impact on AI bot traffic (measured on real WordPress pages):</p>
<ul>
<li><strong>95%+ less bandwidth</strong> per crawled page (typically 15–30x smaller; e.g. a 79 KB page <span aria-hidden="true" class="wp-exclude-emoji">→</span> 4 KB. Heavy page-builder sites see 100x+)</li>
<li><strong>~60–80% fewer database queries</strong> per request</li>
<li><strong>~60% lower peak PHP memory</strong> per request</li>
<li><strong>~50–80% faster Time To First Byte</strong> (measured 54% on a content-heavy page, 67% on a short one)</li>
<li>On a site with 500 posts crawled monthly by major AI bots, that’s roughly <strong>gigabytes <span aria-hidden="true" class="wp-exclude-emoji">→</span> tens of megabytes</strong> of monthly bandwidth and a fraction of the cumulative PHP execution time</li>
</ul>
<p>Numbers vary by theme and content. Heavier setups (Elementor, Divi, Avada) see the biggest savings; lightweight themes see less dramatic but still meaningful gains. The conversion is cached as a transient, so repeated bot hits cost almost nothing.</p>
<h4>How it works</h4>
<ul>
<li>AI crawlers are detected by User-Agent and served Markdown automatically — no configuration required.</li>
<li>Humans, search engines, and unknown bots receive your normal HTML output. ParseLess never affects what real visitors see.</li>
<li>The <code>?format=md</code> query parameter works on any post URL for manual preview or CLI piping.</li>
<li><code>/llms.txt</code> is published at your site root with a list of available content for AI indexers.</li>
<li>Conversion happens once per post and is cached. Subsequent requests serve a single transient read.</li>
</ul>
<h4>Features</h4>
<ul>
<li>Automatic Markdown for known AI bots (GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot, CCBot, Google-Extended, Applebot-Extended, Bytespider, Meta-ExternalAgent, cohere-ai, and more)</li>
<li>Manual preview and CLI access via <code>?format=md</code> on any post URL</li>
<li><code>/llms.txt</code> endpoint for AI-indexing standards</li>
<li>Respects noindex flags from Yoast SEO, Rank Math, and Genesis</li>
<li>Skips private, draft, password-protected, and trashed posts</li>
<li>Works with all public post types (configurable)</li>
<li>Optional YAML frontmatter (title, URL, author, date, categories, tags, excerpt)</li>
<li>Transient-based caching with configurable TTL</li>
<li>Settings page at <strong>Tools <span aria-hidden="true" class="wp-exclude-emoji">→</span> ParseLess</strong> for detection mode, bot list, cache TTL, post types, and llms.txt control</li>
<li>Per-post meta box with Markdown preview and copy-to-clipboard</li>
<li>Extensible via filters: <code>md4ai_bot_list</code>, <code>md4ai_supported_post_types</code>, <code>md4ai_markdown_output</code>, <code>md4ai_cache_ttl</code>, <code>md4ai_should_serve_markdown</code></li>
</ul>
<h3>Privacy and data collection</h3>
<p>When logging is enabled (off by default), ParseLess records the following for<br />
each Markdown request:</p>
<ul>
<li>Post ID, URL, and request timestamp</li>
<li>The full User-Agent header</li>
<li>A salted SHA-256 hash of the requester’s IP address (the raw IP is never stored)</li>
<li>Matched bot identifier, bytes served, and whether the response came from cache</li>
</ul>
<p>IP hashes cannot be reversed from an email address, so the WordPress privacy<br />
exporter/eraser tools will report no personal data on demand.</p>
<p>Logs are pruned daily according to the configured retention window<br />
(7/30/90/365 days, default 30). Site owners can disable logging or click<br />
“Delete all logged requests” at any time from Tools <span aria-hidden="true" class="wp-exclude-emoji">→</span> ParseLess.</p>


ParseLess

Serve your posts as Markdown to AI crawlers and CLI tools — typically 95%+ less bandwidth, ~60–80% fewer DB queries, no theme bloat.

Ambiscale LLMs Friendly vs ParseLess