kkarpieszuk

Kolibia Agent-Ready makes WordPress content available as Markdown so tools and agents do not have to scrape HTML. Behaviour follows common content negotiation patterns (similar in spirit to <a href="https://developers.cloudflare.com/fundamentals/reference/markdown-for-agents/" rel="nofollow ugc">Markdown for Agents</a> on Cloudflare): clients that prefer Markdown can request it explicitly.
What is included today
<ul>
<li>Markdown responses for singular content — For each built-in public post type registered by WordPress core (<code>post</code>, <code>page</code>, <code>attachment</code> on a typical site), a single-item URL can return <code>Content-Type: text/markdown; charset=UTF-8</code> instead of the themed HTML page.</li>
<li>Markdown for the posts index — When the site front page shows latest posts (WordPress “Your homepage displays: Your latest posts”), the homepage and other blog-index URLs return Markdown listing the same posts as the main loop.</li>
<li>Static or WooCommerce shop front page — When a page is set as the front page (including a WooCommerce shop page where WordPress reports <code>is_singular</code> as false), that page’s content is still served as Markdown on <code>/</code> when clients negotiate <code>text/markdown</code>.</li>
<li>When Markdown is served
<ul>
<li>The request includes an <code>Accept</code> header that lists <code>text/markdown</code> (requests with <code>text/markdown;q=0</code> are ignored).</li>
<li>Or the URL includes the query argument <code>output_format=md</code> (handy for browsers and manual testing).</li>
</ul>
</li>
<li>HTML to Markdown — Rendered post body is passed through WordPress’s <code>the_content</code> filters, then converted to Markdown using the <a href="https://github.com/thephpleague/html-to-markdown" rel="nofollow ugc">league/html-to-markdown</a> library. A short YAML front matter block includes <code>title</code> and <code>permalink</code>.</li>
<li>Password-protected content — If a password is required, the response is a small Markdown message instead of the full body.</li>
<li>Discovery in HTML — On the normal HTML view, the plugin prints <code><link rel="alternate" type="text/markdown" href="…?output_format=md" /></code> so clients can find the Markdown URL.</li>
<li>Admin shortcut — On the posts and pages list screens, a row action View as AI Agent opens the same Markdown URL (preview URLs are used for drafts where WordPress would show Preview).</li>
</ul>
Developer filters
<ul>
<li><code>kolibia_ar_markdown_post_types</code> — Defaults to all public, built-in post types from core; override to add custom post types or remove types (e.g. <code>attachment</code>).</li>
<li><code>kolibia_ar_post_markdown</code> — Filter the final Markdown string.</li>
<li><code>kolibia_ar_home_markdown</code> — Filter the Markdown for the posts index / homepage loop.</li>
<li><code>kolibia_ar_home_markdown_post_types</code> — Post types included in the homepage Markdown list (default: <code>post</code>).</li>
<li><code>kolibia_ar_markdown_password_required</code> — Filter the Markdown shown when a password is required.</li>
</ul>
Roadmap
Further “agent readiness” features (broader API discovery, additional formats) may be added in future releases.

Kolibia Agent-Ready

Kolibia Agent-Ready adds API discovery and Markdown responses so AI agents can use your WordPress content and services.

Umay Ajans

Modern AI agents (ChatGPT, Claude, Perplexity, Gemini, etc.) work much better with Markdown than HTML. Umay AI Markdown inspects the incoming <code>Accept</code> header and, only when <code>text/markdown</code> is requested, intercepts the response and serves a clean, agent-friendly Markdown representation of the page.
Browsers, search engines, and any client that does not explicitly ask for Markdown receive the unchanged HTML response. There is no settings page, no cron job, and no external service call.
Key features
<ul>
<li>Zero configuration — install, activate, done.</li>
<li>Only triggers when <code>Accept: text/markdown</code> is present. Regular visitors and search engines are never affected.</li>
<li>Hybrid content extraction: uses <code>the_content</code> for posts/pages, falls back to a DOM-based extractor for archives, taxonomies, and the homepage.</li>
<li>Powered by the industry-standard <a href="https://github.com/thephpleague/html-to-markdown" rel="nofollow ugc">league/html-to-markdown</a> library.</li>
<li>Transient-cached for 12 hours per URL (sha256-keyed). Auto-invalidated on <code>save_post</code>, term edits, theme switches, and menu updates.</li>
<li>Built-in IP rate limiter (30 requests / minute by default) to mitigate abuse.</li>
<li>Strict input sanitization, header injection protection, libxml entity hardening (XXE-safe), and full WordPress Coding Standards compliance.</li>
<li>PSR-4 autoloaded, namespaced OOP code. No globals.</li>
<li>Sends <code>Vary: Accept</code>, <code>X-Robots-Tag: noindex</code>, and <code>X-Content-Type-Options: nosniff</code> on every Markdown response.</li>
</ul>
What gets sent to AI agents
Each Markdown response includes a YAML front-matter block with the page title, site name, canonical URL, and ISO-8601 generation timestamp, followed by the page body converted to Markdown. Navigation, footer, sidebars, scripts, styles, comment forms, related posts, and other page chrome are stripped before conversion.
Filters
Two filters are available for advanced customization:
<ul>
<li><code>umay_mdn_bypass</code> — Return <code>true</code> to skip Markdown handling for the current request.</li>
<li><code>umay_mdn_cache_ttl</code> — Override the default 12-hour cache lifetime (in seconds, minimum 60).</li>
<li><code>umay_mdn_rate_limit</code> — Override the default 30-requests-per-minute rate limit.</li>
<li><code>umay_mdn_converter_options</code> — Modify the league/html-to-markdown converter options array.</li>
</ul>

Umay AI Markdown

Serves clean Markdown to AI agents (ChatGPT, Claude, Perplexity) on Accept: text/markdown. HTML stays default for browsers. Zero-config.

Kolibia Agent-Ready vs Umay AI Markdown