內容簡介
Msite AI Crawler Monitor 外掛能夠幫助網站管理者了解 AI 爬蟲如何訪問他們的網站，提供詳細的爬蟲統計數據和 AI 健康檢查，確保網站對 AI 引擎的友好性。
【主要功能】
• 爬蟲統計：記錄多種知名 AI 爬蟲的訪問情況
• 分類掃描：區分學習掃描與即時查詢
• AI 健康檢查：檢查網站的基本 AI 兼容性
• 期間篩選：支持多種時間範圍的數據過濾
• 隱私保護：不儲存 IP 地址或完整用戶代理字串
外掛標籤
開發者團隊
原文外掛簡介
Msite AI Crawler Monitor shows you how AI engines treat your website.
Bot statistics
Logs visits from known AI crawlers: GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-SearchBot, Claude-User, PerplexityBot, Perplexity-User, Google-Extended, GoogleOther, Bytespider, Amazonbot and Meta’s external agent.
Separates learning scans (crawlers that build an index or training set) from real-time queries (requests sent the moment a person asks an AI a question).
Period filter (7 days, 30 days, 3 months, year, all time), per-bot table, donut charts and the top pages visited by bots.
AI Health check
Checks your live site for the basics AI engines rely on: HTTPS, robots.txt (and whether it blocks AI bots), XML sitemap, search-engine visibility setting, llms.txt, Schema JSON-LD and canonical tag. Each check is shown as a traffic-light card with a short explanation.
Pro add-on (separate, optional)
A separate Pro add-on adds an AI-readable content layer (Quick Facts, FAQ, schema), FAQ drafting with your own AI API key, Google Analytics 4 AI referrals and a readiness score. It is a different plugin, licensed personally for now. This plugin works fully without it and contains none of its code.
Hebrew first
The plugin is fully translatable (Hebrew translation is available through translate.wordpress.org) and uses logical CSS so it looks right in both RTL and LTR admin screens.
External services
This plugin does not send data to any external service. The AI Health check only makes HTTP requests to your own site (home page, robots.txt, sitemap and llms.txt). Links shown next to check results (for example to Google’s documentation) are plain links that open only when you click them.
Privacy
The plugin stores only a daily counter per bot and page (date, bot name, post ID, number of visits). It does not store IP addresses or full user-agent strings, and it does not log logged-in users.
Bots are recognised by the name in their User-Agent header, which any client can fake, so treat the numbers as indicative.
Third-party libraries
Chart.js 4.5.1 (MIT License), bundled locally in assets/vendor/. Source: https://www.chartjs.org
