gregrandall

AI agents, LLMs, and crawlers have to wade through navigation bars, sidebars, ads, and comment forms to reach the content they want, and every element costs tokens. <a href="https://blog.cloudflare.com/markdown-for-agents/" rel="nofollow ugc">Cloudflare measured</a> an 80% reduction in token usage when converting a blog post from HTML to Markdown (16,180 tokens down to 3,150).
Botkibble adds a Markdown endpoint to every published post and page.
Cloudflare offers <a href="https://developers.cloudflare.com/fundamentals/reference/markdown-for-agents/" rel="nofollow ugc">Markdown for Agents</a> at the CDN edge on Pro, Business, and Enterprise plans. Botkibble does the same thing (for free) at the origin, so it works on any host.
<a href="https://github.com/greg-randall/botkibble" rel="nofollow ugc">GitHub Repository</a>
Three ways to request Markdown:
<ul>
<li><code>.md</code> suffix: append <code>.md</code> to any post or page URL (e.g. <code>example.com/my-post.md</code>)</li>
<li>Query parameter: add <code>?format=markdown</code> to any post or page URL</li>
<li>Content negotiation: send <code>Accept: text/markdown</code> in the request header</li>
</ul>
What’s in every response:
<ul>
<li>Structured metadata header with title, date, categories, tags, word count, character count, and estimated token count (in YAML frontmatter format, readable by any AI agent)</li>
<li>Clean Markdown converted from fully-rendered post HTML (shortcodes run, filters applied)</li>
<li><code>Content-Type: text/markdown</code> and <code>Vary: Accept</code> response headers</li>
<li><code>Content-Signal</code> header for AI signal declaration — defaults to <code>ai-train=no, search=yes, ai-input=yes</code> — see <a href="https://contentsignals.org/" rel="nofollow ugc">contentsignals.org</a></li>
<li><code>X-Markdown-Tokens</code> header with estimated token count</li>
<li>Discovery via <code><link rel="alternate"></code> in the HTML head and <code>Link</code> HTTP header</li>
<li>Automatic cache invalidation when a post is updated or deleted</li>
</ul>
Performance:
Botkibble writes Markdown to disk on the first request, then serves it as a static file. A built-in Fast-Path serves cached files during WordPress’s <code>init</code> hook, before the main database query runs. No extra configuration needed.
Add a web server rewrite rule and Botkibble bypasses PHP entirely, serving <code>.md</code> files the same way a server would serve an image or CSS file:
 Method 
 Avg. response time
 Standard HTML 
 0.97s
 Markdown (cold, first request) 
 0.95s
 Markdown (cached, PHP Fast-Path) 
 0.87s
 Markdown (Nginx/Apache direct) 
 0.11s
Serving directly from disk is 88% faster than a full WordPress page load. See the Performance section below for Nginx and Apache configuration.
Security:
<ul>
<li>Drafts, private posts, and password-protected content return <code>403 Forbidden</code></li>
<li>Rate limits cache-miss regenerations (20/min by default) to mitigate DoS abuse</li>
<li><code>X-Robots-Tag: noindex</code> keeps Markdown versions out of search results</li>
<li><code>Link: rel="canonical"</code> points search engines back to the HTML version</li>
</ul>
Cache variants (optional):
You can persist alternate cached representations by adding <code>?botkibble_variant=slim</code> (or any other variant name). 
Variant caches are stored under:
<pre><code>/wp-content/uploads/botkibble/_v/<variant>/<slug>.md
</code></pre>
What it does NOT do:
<ul>
<li>Expose drafts, private posts, or password-protected content</li>
<li>Serve non-post/page content types by default</li>
<li>Require any configuration. Activate and it works.</li>
</ul>
<h3>Why Markdown?</h3>
HTML is expensive for AI systems to process. <a href="https://blog.cloudflare.com/markdown-for-agents/" rel="nofollow ugc">Cloudflare measured</a> an 80% reduction in token usage when converting a blog post from HTML to Markdown (16,180 tokens down to 3,150).
Cloudflare now offers <a href="https://developers.cloudflare.com/fundamentals/reference/markdown-for-agents/" rel="nofollow ugc">Markdown for Agents</a> at the CDN edge via the <code>Accept: text/markdown</code> header, available on Pro, Business, and Enterprise plans.
This plugin does the same thing at the origin, so it works on any host. It also adds <code>.md</code> suffix URLs, <code>?format=markdown</code> query parameters, YAML frontmatter, static file caching, and server-level offloading.
If you use Cloudflare, both share the same <code>Accept: text/markdown</code> header, <code>Content-Signal</code> headers, and <code>X-Markdown-Tokens</code> response headers.
Cloudflare currently defaults to <code>Content-Signal: ai-train=yes, search=yes, ai-input=yes</code> with no way to change it. Botkibble defaults to <code>ai-train=no</code> and lets you override the full signal per site via the <code>botkibble_content_signal</code> filter.
<h3>Performance & Static Offloading</h3>
This plugin supports static file offloading by writing Markdown content to <code>/wp-content/uploads/botkibble/</code>.
<h3>Nginx Configuration</h3>
To bypass PHP entirely and have Nginx serve the files (including variants) directly:
<pre><code># Variants
location ~* ^/(_v/[^/]+/.+)\.md$ {
 default_type text/markdown;
 try_files /wp-content/uploads/botkibble/$1.md /index.php?$args;
}

# Default
location ~* ^/(.+)\.md$ {
 default_type text/markdown;
 try_files /wp-content/uploads/botkibble/$1.md /index.php?$args;
}
</code></pre>
<h3>Apache (.htaccess)</h3>
Add this to your <code>.htaccess</code> before the WordPress rules:
<pre><code>RewriteEngine On
# Variants
RewriteCond %{DOCUMENT_ROOT}/wp-content/uploads/botkibble/_v/$1/$2.md -f
RewriteRule ^_v/([^/]+)/(.+)\.md$ /wp-content/uploads/botkibble/_v/$1/$2.md [L,T=text/markdown]

# Default
RewriteCond %{DOCUMENT_ROOT}/wp-content/uploads/botkibble/$1.md -f
RewriteRule ^(.*)\.md$ /wp-content/uploads/botkibble/$1.md [L,T=text/markdown]
</code></pre>
Even without these rules, the plugin uses a “Fast-Path” that serves cached files from PHP before the main database query is executed.
<h3>Credits</h3>
We thank Cristi Constantin (https://github.com/cristi-constantin) for contributing cache variants, URL and SEO improvements, and fixing important bugs.

Botkibble

Serves every published post and page as Markdown for AI agents and crawlers. No configuration, no API keys. Activate and it works.

Ayuda WordPress

Fernando Tellado

AI Content Signals lets you declare, in a machine-readable way, how AI systems may use your content: for search indexing, real-time AI answers (RAG), or model training. It started with Cloudflare’s Content Signals in robots.txt and now expresses the same preferences across several surfaces, so more crawlers and tools can read them:
<ul>
<li>robots.txt — Cloudflare Content Signals (search / ai-input / ai-train)</li>
<li>HTTP header — <code>X-Robots-Tag: noai, noimageai</code></li>
<li>HTML meta — robots <code>noai, noimageai</code></li>
<li>/.well-known/tdmrep.json — W3C Text and Data Mining Reservation Protocol</li>
<li>EU Directive 2019/790 rights reservation</li>
</ul>
Everything is opt-in and does not change your robots.txt output unless you enable it.
The signals you control
You set three preferences, and AI Content Signals expresses each one in the right format on every surface you enable:
<ul>
<li>search — allow or deny search indexing and traditional search results (links and short snippets)</li>
<li>ai-input — allow or deny using your content for real-time AI answers (RAG, grounding, AI Overviews)</li>
<li>ai-train — allow or deny using your content to train or fine-tune AI models</li>
</ul>
These three come from Cloudflare’s Content Signals vocabulary, written to your robots.txt. When you deny AI training, the same opt-out is also emitted on any other surface you enable: <code>noai, noimageai</code> in an HTML meta tag and an <code>X-Robots-Tag</code> header, plus a <code>tdm-reservation</code> in your TDMRep manifest. Declaring the same preference in several places means a crawler that ignores one signal may still honor another.
Key Features
<ul>
<li>Easy-to-use settings page in WordPress admin</li>
<li>Set global defaults for all crawlers</li>
<li>Configure specific settings for individual AI bots (GPTBot, ClaudeBot, PerplexityBot, etc.)</li>
<li>Add custom bot User-Agents</li>
<li>Supports both physical and virtual robots.txt files</li>
<li>Option to create physical robots.txt with basic WordPress rules</li>
<li>Preview generated Content Signals before applying</li>
<li>Export and import settings as JSON for easy migration between sites</li>
<li>Optional legal text with EU Directive reference</li>
<li>Developer-friendly: filter hook to extend the predefined bots list</li>
<li>Works with existing robots.txt from SEO plugins</li>
<li>Automatic sitemap detection and inclusion</li>
<li>Optional extra output surfaces: HTML robots meta tag (noai, noimageai), X-Robots-Tag header, and a W3C TDMRep file at /.well-known/tdmrep.json</li>
</ul>
Supported Bots
The plugin includes predefined settings for 28 major AI crawlers:
<ul>
<li>OpenAI GPTBot, OAI-SearchBot, and ChatGPT-User</li>
<li>Anthropic ClaudeBot, Claude-Web, and anthropic-ai</li>
<li>Perplexity Bot and Perplexity-User</li>
<li>Google Extended (Gemini) and GoogleOther</li>
<li>Amazon Bot</li>
<li>Apple Extended</li>
<li>Meta/Facebook Bot and meta-externalagent</li>
<li>DuckDuckGo DuckAssistBot</li>
<li>Allen Institute AI2Bot</li>
<li>Mistral AI</li>
<li>ByteDance Bytespider</li>
<li>DeepSeek AI</li>
<li>xAI Grok</li>
<li>Huawei Pangu</li>
<li>Common Crawl, Cohere AI, Diffbot, You.com Bot, and more</li>
</ul>
Important Notice
Content Signals is a declarative standard – it expresses your preferences but does not technically enforce them. AI companies are not legally required to respect these signals, though the plugin includes legal text referencing EU copyright directives.
The IETF AI Preferences (AIPREF) Working Group is currently developing a formal standard based on similar concepts. This plugin implements the current Cloudflare Content Signals specification and will be updated as standards evolve.
This plugin works best when combined with other protection measures like traditional robots.txt rules and server-level bot management.
<h3>Support</h3>
Need private support or custom development?
Do you need one-on-one help, priority troubleshooting, or a custom feature, integration, or tweak built specifically for your site? I offer private support and custom development. Just <a href="mailto:ai-content-signals@ayudawp.com" rel="nofollow ugc">contact me</a> and tell me what you need.
Need help or have suggestions?
<ul>
<li><a href="https://servicios.ayudawp.com" rel="nofollow ugc">Official website</a></li>
<li><a href="https://wordpress.org/support/plugin/ai-content-signals/" rel="ugc">WordPress support forum</a></li>
<li><a href="https://www.youtube.com/AyudaWordPressES" rel="nofollow ugc">YouTube channel</a></li>
<li><a href="https://ayudawp.com" rel="nofollow ugc">Documentation and tutorials</a></li>
</ul>
Love the plugin? Please leave us a <a href="https://wordpress.org/support/plugin/ai-content-signals/reviews/#new-post" rel="ugc">5-star review</a> and help spread the word!
<h3>About AyudaWP.com</h3>
We are specialists in WordPress security, SEO, and performance optimization plugins. We create tools that solve real problems for WordPress site owners while maintaining the highest coding standards and accessibility requirements.

AI Content Signals

Control how AI uses your content — search, AI answers or training — via robots.txt Content Signals, HTTP headers, meta tags and TDMRep.

Botkibble vs AI Content Signals