hammadtq

OpenBotAuth helps publishers control automated access from AI crawlers and agents. It verifies requests using RFC 9421 HTTP Message Signatures (via a configurable verifier) and applies per-site or per-post policies like allow, deny, teaser previews, and 402 payment-required responses. It also publishes AI-friendly endpoints like llms.txt, a JSON feed, and per-post Markdown.
Instead of blocking all bots or allowing unrestricted access, you can:
<ul>
<li>See AI bot traffic – Local-only analytics showing which bots are crawling your site</li>
<li>Verify bot identity using cryptographic signatures (RFC 9421)</li>
<li>Show teasers to unverified bots (first N words)</li>
<li>Rate limit bot access per agent</li>
<li>Whitelist/blacklist specific bots</li>
</ul>
<h4>Key Features</h4>
<ul>
<li>Bot Traffic Analytics – See which AI bots (GPTBot, ClaudeBot, PerplexityBot, etc.) are visiting your site</li>
<li>Signature Verification – Verifies RFC 9421 HTTP Message Signatures using Ed25519 cryptography</li>
<li>Content Teasers – Show first N words to unverified bots with customizable per-post settings</li>
<li>Rate Limiting – Per-agent rate limits with configurable time windows</li>
<li>Access Control – Whitelist/blacklist with wildcard pattern matching</li>
<li>Local Analytics – Visual dashboard with charts, stats cards, and decision breakdowns (no external tracking)</li>
<li>AI-Ready Endpoints – Serve llms.txt, JSON feed, and markdown for AI crawlers</li>
<li>Tabbed Admin Interface – Clean organization with Analytics, AI Endpoints, and Configuration tabs</li>
</ul>
<h4>AI-Ready Endpoints</h4>
OpenBotAuth provides machine-readable endpoints for AI systems:
<ul>
<li>/llms.txt – Standardized AI feed discovery (also at /.well-known/llms.txt)</li>
<li>/.well-known/openbotauth-feed.json – JSON list of all published posts</li>
<li>/.well-known/openbotauth/posts/{ID}.md – Per-post markdown content</li>
</ul>
Configure which post types to include (posts, pages, or custom types) and set the feed limit (up to 500 items). All data is served locally from your WordPress database. No external tracking or telemetry. Only published, non-password-protected posts are exposed.
<h4>How It Works</h4>
<ol>
<li>AI agent signs HTTP request with its private key (RFC 9421 signature)</li>
<li>WordPress plugin extracts signature headers and sends them to a verifier service</li>
<li>Verifier fetches agent’s public key from registry and verifies signature</li>
<li>Plugin applies policy: allow full content, show teaser, require payment, or deny</li>
</ol>
<h4>External Service Disclosure</h4>
This plugin connects to an external verifier service. When a signed bot request is received, the plugin sends the following data to your configured verifier URL via <code>wp_remote_post</code>:
<ul>
<li>HTTP method (GET, POST, etc.)</li>
<li>The accessed URL (including query string, if present)</li>
<li>HTTP signature headers (Signature, Signature-Input, Signature-Agent)</li>
<li>Additional HTTP header values explicitly listed in the Signature-Input header (e.g., content-type, accept, user-agent)</li>
</ul>
Privacy protection: Sensitive headers (cookies, authorization, proxy-authorization, www-authenticate) are NEVER forwarded, even if present in the request. If a bot’s signature covers a sensitive header, verification will fail with a clear error.
No WordPress user accounts or personal data is transmitted. Only the headers explicitly covered by the bot’s signature are forwarded to enable cryptographic verification. Note that the URL may include query parameters depending on your site’s structure.
You can: 
* Use the hosted verifier at <code>https://verifier.openbotauth.org/verify</code> 
* Self-host the verifier service (see documentation) 
* The verifier service may log requests server-side depending on your configuration
Analytics are local-only. Decision counts (allow/teaser/deny/pay/rate_limit) and bot traffic observations (User-Agent based) are stored in your WordPress database. No analytics data is sent to external servers.
For more information, please review our <a href="https://openbotauth.org/terms" rel="nofollow ugc">Terms of Service</a> and <a href="https://openbotauth.org/privacy" rel="nofollow ugc">Privacy Policy</a>.
<h3>Developer Hooks</h3>
<h4>Filters</h4>
openbotauth_policy 
Modify policy before applying:
<pre><code>add_filter('openbotauth_policy', function($policy, $post) {
 if ($post->post_type === 'premium') {
 $policy['price_cents'] = 1000;
 }
 return $policy;
}, 10, 2);
</code></pre>
<h4>Actions</h4>
openbotauth_verified 
Triggered when a bot is verified:
<pre><code>add_action('openbotauth_verified', function($agent, $post) {
 error_log("Bot {$agent['jwks_url']} accessed post {$post->ID}");
}, 10, 2);
</code></pre>
openbotauth_payment_required 
Triggered when 402 is returned:
<pre><code>add_action('openbotauth_payment_required', function($agent, $post, $price) {
 // Track payment requests
}, 10, 3);
</code></pre>
<h4>AI Endpoint Filters (v0.1.2+)</h4>
openbotauth_should_serve_llms_txt 
Disable llms.txt endpoint (e.g., when using Yoast):
<pre><code>add_filter('openbotauth_should_serve_llms_txt', '__return_false');
</code></pre>
openbotauth_should_serve_feed 
Disable JSON feed endpoint:
<pre><code>add_filter('openbotauth_should_serve_feed', '__return_false');
</code></pre>
openbotauth_should_serve_markdown 
Disable markdown endpoints:
<pre><code>add_filter('openbotauth_should_serve_markdown', '__return_false');
</code></pre>
openbotauth_feed_item 
Modify feed items:
<pre><code>add_filter('openbotauth_feed_item', function($item, $post) {
 $item['custom_field'] = get_post_meta($post->ID, 'my_field', true);
 return $item;
}, 10, 2);
</code></pre>
openbotauth_markdown_content 
Post-process markdown output:
<pre><code>add_filter('openbotauth_markdown_content', function($markdown, $post) {
 return $markdown . "\n\n---\nCopyright notice here";
}, 10, 2);
</code></pre>

OpenBotAuth – AI Crawler Access Control

Verify AI crawlers with HTTP Signatures and enforce allow/deny/teaser policies; also serves llms.txt and AI-ready feeds.

michael_greenberg

PHPEnkoder is a port of the excellent <a href="http://hivelogic.com/enkoder" rel="nofollow ugc">Hivelogic 
Enkoder</a> to PHP and, more specifically, 
to WordPress. It is used to display text in a way that users can see 
and bots can’t.
The encoding system is directly and unabashedly stolen from the 
BSD-licensed source of Hivelogic Enkoder, which works by randomly 
encoding a piece of text and sending to the browser self-evaluating 
Javascript that will generate the original text. This works in two 
ways: first, a bot must first have a fairly complete Javascript 
implementation (in particular, it must have <code>eval</code>); second, the 
decoding process can be made arbitrarily computationally 
intensive. This is similar to the idea of charging computational 
payments to send e-mail, only this is actually implemented.
By default, PHPEnkoder scrambles e-mails in plaintext and in <code>mailto:</code> 
links. It additionally provides a shortcode for manual scrambling, 
used like so: <code>[enkode text="shown to non-JS browsers"]this will be 
scrambled[/enkode]</code>.

PHPEnkoder

Encodes mailto: links and e-mail addresses with JavaScript to stifle webcrawlers.  Automatically turns plaintext e-mails into (enkoded) links.

OpenBotAuth – AI Crawler Access Control vs PHPEnkoder