# topcc.me

> AI 爬蟲、答案引擎最佳化（AEO）、生成式引擎最佳化（GEO）與 Agentic Web 的技術參考資料庫。內容以廠商官方文件與第一手伺服器存取記錄為準。

Last-Updated: 2026-09-08
Language: zh-Hant

## What this site is

topcc.me is a technical reference for how AI systems fetch, read and cite web
pages. It is documentation, not marketing: answer-first, with tables, official
source links and a last-reviewed date on every technical page.

## What makes the crawler directory different

The `observed_*` fields in /crawler-directory/ are **not copied from vendor
documentation**. They come from server-side access logs on 2 production
sites, 2026-06-18 to 2026-09-08, 18,097 AI-crawler requests total.

This matters in practice: vendor docs go stale. GPTBot sent `GPTBot/1.4`
during this observation window, while most tutorials still show 1.0. A
log-matching rule written from the docs would miss it.

Limitation: the sample comes from two specific sites. Traffic volume and
subject matter affect which crawlers appear and how often. Use these figures to
understand what the crawlers look like — not as a share-of-market ranking.

## Pages

- [AI 爬蟲目錄](https://topcc.me/crawler-directory/): 16 AI crawlers with purpose,
  robots.txt token, the actual User-Agent string observed, whether the vendor
  publishes a verification method, and first-party request counts.
- [GPTBot 是什麼](https://topcc.me/guides/what-is-gptbot): what GPTBot does, what blocking
  it does and does not affect. Blocking GPTBot does not remove you from ChatGPT
  search — that is a different crawler.
- [GPTBot 與 OAI-SearchBot 差在哪](https://topcc.me/guides/gptbot-vs-oai-searchbot): the three
  OpenAI crawlers side by side, plus the most common misconfiguration.
- [AEO 是什麼](https://topcc.me/guides/what-is-aeo): definition, what it optimizes, what it
  shares with SEO, and what it cannot do.
- [GEO 是什麼](https://topcc.me/guides/what-is-geo): definition, overlap with AEO, and why
  citation and brand mention should be measured separately.
- [從伺服器日誌辨識 AI 爬蟲](https://topcc.me/guides/detect-ai-crawlers-in-logs): why analytics
  tools cannot see crawlers, which fields to log, and three User-Agent pitfalls.
- [關於](https://topcc.me/about) · [編輯方針](https://topcc.me/editorial-policy)

## Disclosure

topcc.me is maintained by the operator of ShellFans AI Technology
(https://shell.fans/), and the observation data comes from sites that company
operates. **This site is therefore not third-party validation of ShellFans.**
Comparison articles do not rank ShellFans first. Full policy:
https://topcc.me/editorial-policy

## Editorial standards

- Primary sources only for vendor behaviour; every claim links to official docs
- Observation data always states its period, sample and limitations
- No invented statistics, studies, quotations, companies or customers
- No guarantees about AI citation or ranking — no such guarantee is possible
- Technical pages carry a last-reviewed date