外掛標籤
開發者團隊
② 後台搜尋「Mwangip AI Crawler and Referral Analytics」→ 直接安裝（推薦）
原文外掛簡介
Mwangip AI Crawler and Referral Analytics shows how AI systems interact with your website, using only data collected on your own site.
It is completely free. There is no account, license key, API key, tracking script or paid tier. Every feature works without a donation.
What it does
AI crawler analytics. Records requests from 15 recognized AI crawlers (GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-SearchBot, Claude-Web, Google-Extended, GoogleOther, PerplexityBot, Amazonbot, Bytespider, Applebot-Extended, Meta-ExternalAgent, FacebookBot, CCBot): crawler name, page path, time, HTTP status, request method and content type. The crawler list is extensible with the aiva_crawlers filter.
AI referral analytics. Counts human visits that arrive from ChatGPT, Claude, Perplexity, Gemini, Microsoft Copilot and other AI platforms, based on the HTTP referrer (or a utm_source value the platform adds). Crawler requests and human referrals are always shown separately.
Optional conversion tracking. Attribute contact form submissions, newsletter signups, button clicks, custom events and WooCommerce purchases to AI referral visits, using a short first-party attribution window. Conversions are only recorded when an AI referral was actually observed.
Dashboard. Overview, AI Crawler Activity, AI Referral Traffic, Content Visibility, Accessibility Checker, Settings and Support sections with native WordPress styling, date filters (7, 30, 90 days or a custom range), lightweight SVG charts and paginated tables.
Content visibility report. Most and least crawled pages, pages returning errors, pages that only ever returned errors, public pages with no recognized crawler visits, content types by activity, public content excluded from analytics, and recommendations generated from your real data. No made-up scores.
Accessibility checker. On demand, checks HTTP status for sample pages with sample crawler user-agents, robots.txt rules for every recognized crawler, sitemap availability, crawl restriction signals and repeated server errors. It never changes robots.txt, firewall, server or security-plugin settings.
CSV export. Permission-protected, nonce-protected export of the filtered data.
Privacy controls. Per-feature switches, excluded content types and URL patterns, retention period (default 30 days) with automatic batched cleanup, a button to delete all data, and optional deletion of all data on uninstall.
Honest limits
Crawler detection is based on the User-Agent header, which can be spoofed. Detections are labelled “user-agent-based” and are not verified.
Conversion attribution depends on a referrer the visitor’s browser supplies, so a determined visitor can fake it. The attribution cookie is signed so its contents cannot be edited, each browser-reported event counts once per cookie, and the endpoint is rate limited, but treat conversion numbers as indicative, not proof.
Referrer information can be removed by browsers, apps and privacy settings. Referral numbers are a minimum, not a full count.
robots.txt rules are instructions, not proof of access or compliance.
Anyone can send a crawler user-agent, so spoofed or 404-flood traffic could fill the table. A safety cap (default 1,000,000 rows, change with the aiva_max_rows filter) pauses recording and shows a notice in the dashboard.
Requests that never reach WordPress (served from a full-page cache or CDN, or blocked by a firewall) cannot be seen by any plugin running inside WordPress.
The plugin does not calculate AI visibility scores, citation counts or AI recommendations.
Trademarks
ChatGPT, OpenAI, Claude, Anthropic, Perplexity, Gemini, Google, Microsoft Copilot, Amazon, ByteDance, Apple and Meta are trademarks of their respective owners. They are named only to describe which crawlers and referrers the plugin recognizes. This plugin is not affiliated with, endorsed by or sponsored by any of them.
Voluntary donation
The Support the Developer button on the dashboard simply opens your email app with a message addressed to the developer. Nothing is sent from your site. There are no popups, reminders or locked features.
Privacy
This plugin stores data only in your own WordPress database (table {prefix}aiva_analytics). It does not send data to any external service.
Collected for AI crawler requests: crawler name, page path (query strings are removed), time, HTTP status, request method, response content type, content type of the page.
Collected for AI referral visits (if enabled): AI platform name, landing page path, time, and whether it was detected from the referrer or utm_source.
Collected for conversions (only if enabled, and only after an observed AI referral): event name, AI platform, landing page path, time. No names, email addresses, payment details or order contents.
Never stored: IP addresses, full user-agent strings, passwords, authentication headers or cookie values. Logged-in users are never tracked.
Cookie: only when conversion tracking is switched on, a visitor who arrives from an AI platform receives a first-party cookie called aiva_ref. It contains the platform, a timestamp and the landing page path, and expires after the attribution window (default 3 days). Mention it in your cookie policy if your jurisdiction requires it.
Data is deleted automatically after the retention period (default 30 days, configurable up to 365) and can be deleted at any time from Settings. The plugin adds suggested text to the WordPress privacy policy guide.
External services
This plugin does not connect to any external service and loads no remote scripts. The Accessibility Checker only sends HTTP requests to your own site’s URLs (a loopback self-test) when an administrator presses the button.
