ABNet WP-PostViews Addons

A lightweight extension for WP-PostViews that blocks view counting for known bots and social/embed traffic, based on rules from data/bots.json and optional strict referer-domain matching.

Features

• Extends WP-PostViews through its native postviews_should_count filter;
• Blocks view count for requests by User-Agent patterns from data/bots.json (bots section);
• Blocks view count for requests by Referer patterns from data/bots.json (referers section);
• Supports strict referer-domain matching (exact domain or subdomain only);
• Includes a Tools page for updating the local bots list from ai-robots-txt;
• Supports fallback to the last downloaded source when remote download fails;
• Includes a tab that previews current bots.json sections in admin.

How It Works

The plugin hooks into postviews_should_count and returns false for requests that match:

• bot User-Agent patterns;
• referer patterns.

In strict mode, referer checks are host/domain-aware instead of simple substring checks.

Requirements

• WordPress 6.x;
• PHP 8.0+;
• WP-PostViews plugin active;
• PHP cURL extension recommended for downloading bots updates.

Install

1. Copy the plugin folder to wp-content/plugins/abnet-wp-post-views-addons.
2. Activate the plugin from WordPress Admin.
3. Ensure WP-PostViews is active and configured.
4. (Optional) Go to Tools > ABNet Post Views Addons to update bots data.

Configurable Constants

Define constants in wp-config.php before WordPress loads plugins.

Constant	Type	Default	Purpose
ABNET_WP_POST_VIEWS_STRICT	bool	false (when undefined)	Enables strict referer host matching (example.com and *.example.com).
ABNET_WP_POST_VIEWS_TRACKING_ENABLED	bool	true (when undefined)	Enables storing tracked visit metadata. Set to false to keep bot exclusion behavior without saving visit rows.
ABNET_WP_POST_VIEWS_FUZZY_BOT_SCORE_THRESHOLD	int	65 (when undefined, non-int, or <= 0)	Minimum fuzzy bot score that auto-flags a request as bot after score calculation.
ABNET_WP_POST_VIEWS_AI_ROBOTS_TXT_URL	string	https://raw.githubusercontent.com/ai-robots-txt/ai.robots.txt/main/robots.json	Overrides the source URL used by the updater.

Example

define('ABNET_WP_POST_VIEWS_STRICT', true);
define('ABNET_WP_POST_VIEWS_TRACKING_ENABLED', true);
define('ABNET_WP_POST_VIEWS_FUZZY_BOT_SCORE_THRESHOLD', 65);
define('ABNET_WP_POST_VIEWS_AI_ROBOTS_TXT_URL', 'https://raw.githubusercontent.com/ai-robots-txt/ai.robots.txt/main/robots.json');

Plugin Filters

abnet_wpv_addons_bot_patterns

Filters bot User-Agent patterns loaded from bots.json.

• Input: string[] $defaultBotPatterns
• Output: string[]

add_filter('abnet_wpv_addons_bot_patterns', function(array $patterns): array {
    $patterns[] = 'MyCustomCrawler';
    return array_values(array_unique($patterns));
});

abnet_wpv_addons_referer_patterns

Filters referer patterns loaded from bots.json.

• Input: string[] $defaultPatterns
• Output: string[]

add_filter('abnet_wpv_addons_referer_patterns', function(array $patterns): array {
    $patterns[] = 'my-embed-host.example';
    return array_values(array_unique($patterns));
});

abnet_wpv_addons_normalized_domain

Filters a normalized referer domain before strict matching. A null return value will cause the domain to be skipped altogether.

• Input: string $normalizedDomain
• Output: string|null

add_filter('abnet_wpv_addons_normalized_domain', function(string $domain): string {
    return trim(strtolower($domain));
});

abnet_wpv_addons_last_update_message

Filters the info message shown on the Tools page after successful updates.

• Input:
  • string $lastUpdateMessage
  • string $userDisplayName
  • string $formattedDate
• Output: string

add_filter('abnet_wpv_addons_last_update_message', function(
    string $message,
    string $user,
    string $date
): string {
    return sprintf('Bots list was last refreshed on %s by %s.', $date, $user);
}, 10, 3);

abnet_wpv_addons_fuzzy_bot_score

Filters the computed bot probability score saved with each tracked visit.

• Input:
  • ABNetWpPostViewsAddons\FuzzyBotScore $score
  • ABNetWpPostViewsAddons\VisitInfo $visitInfo
• Output: ABNetWpPostViewsAddons\FuzzyBotScore

add_filter('abnet_wpv_addons_fuzzy_bot_score', function(
    ABNetWpPostViewsAddons\FuzzyBotScore $score,
    ABNetWpPostViewsAddons\VisitInfo $visitInfo
): ABNetWpPostViewsAddons\FuzzyBotScore {
    if ($visitInfo->getCfIpCountry() === 'RO') {
        return new ABNetWpPostViewsAddons\FuzzyBotScore(
            $score->getScore() - 5,
            $score->getSignals()
        );
    }

    return $score;
}, 10, 2);

Fuzzy Bot Score

Tracked visits include a fuzzy_bot_score value from 0 to 100, estimating how likely the request is to be automated after deterministic bots.json matching has already happened. The calculator returns a FuzzyBotScore object containing both the normalized score and the contributing signal map.

The fuzzy score is based only on VisitInfo fields, not on the already-computed bot/referer matching flags. It uses softer request-shape signals such as user-agent structure, missing or weak Accept, missing Accept-Language, suspicious network names, unusual request methods, no cookies, protocol hints, referer shape, search terms, and UTM parameters. Normal browser signals, local referers, search referers, HTTP/2 or HTTP/3, rich browser Accept headers, and campaign-tagged visits can reduce the score slightly.

Current signals include:

• suspicious or missing User-Agent values, including malformed Safari-like strings, old Chrome versions, and abandoned Windows desktop masks;
• missing, wildcard, or non-HTML Accept headers;
• missing Accept-Language;
• request methods other than GET or HEAD;
• protocol hints such as HTTP/1.0, HTTP/2, and HTTP/3;
• missing, local, or search referers;
• missing cookies;
• search and UTM campaign hints;
• invalid Cloudflare country metadata and suspicious hosting/cloud network names.

bots.json Structure

The plugin reads data/bots.json. Typical structure:

{
    "bots": [
        "Googlebot",
        "bingbot",
        "facebookexternalhit"
    ],
    "referers": [
        "facebook.com",
        "l.facebook.com",
        "t.co",
        "linkedin.com"
    ]
}

Tools Page

Under Tools > ABNet Post Views Addons:

• Update tab:

  • downloads latest source and merges only the bots section;
  • can use last downloaded file as fallback.

• Current bots.json tab:

  • displays all current sections and values from local data/bots.json.

Screenshots

Tools - Update tab

Tools - Current bots tab

Support Script (support/test.php)

This repository also contains a CLI helper script for manual traffic simulation and diagnostics.

Note: this script is for development/support workflows and is not distributed with the plugin package.

Usage:

php test.php <url> [--limit=NUMBER] [--auto-batch] [--no-delay] [--legit-requests]

Parameters:

• <url>: Required. Full target URL (example: https://example.com/article).
• --limit=NUMBER: Optional. Maximum number of scenarios to run from the beginning. Use 0 (default) to run all scenarios from bots.json.
• --auto-batch: Optional. Skips confirmation prompt between batches.
• --no-delay: Optional. Disables random delay (1-2s) between requests.
• --legit-requests: Optional. Emits legitimate-looking browser requests with random normal/search referers, required Accept and Accept-Language headers, and optional Cloudflare country/AS headers. If no positive --limit is provided, 50 requests are sent.

Examples:

php test.php "https://example.com/article" --limit=25
php test.php "https://example.com/article" --auto-batch --no-delay
php test.php "https://example.com/article" --legit-requests --limit=50

Notes

• The plugin respects WP-PostViews native "Exclude Bot Views" setting.
• If WP_DEBUG is enabled, diagnostic logs may be written using error_log.

Version History

Version	Changes
1.0.0	Initial release with bot/referer filtering, strict mode, updater, and Tools integration.