Bytespider

ByteDance’s crawler, gathering content used for TikTok’s search and for Doubao.

Measured across 58 domains read on 12 October 2026, against 13.5% on 2 August 2026.

Check one domainAll eight agents
How the corpus answers Bytespiderread 2026-10-12
5.2%of sites
disallow it
Blocked 5.2%Allowed by name 1.7%Not named 94.8%
2026-08-023.5% – 18.1%2026-10-12
Block rate over 72 recorded days. The denominator is stored with each point, so the line is not restated when the corpus grows.
Operated byByteDance
Blocked by5.2%3 domains
Allowed by name1.7%the file names it
Not named94.8%the file is silent
Sample58domains read
robots.txt tokenBytespidercase-insensitive

What blocking Bytespider costs

Your content stops being gathered for ByteDance products.

What it does not cost

It has no relationship to any Western search engine.

It has a long-standing reputation for crawling aggressively, so some sites block it for load reasons rather than for anything to do with AI.

ByteDance’s own documentation →

How to block itRFC 9309
User-agent: Bytespider
Disallow: /
User-agent: Bytespider
Disallow:

Allowed, and said out loud.

To allow it and say so, use a bare Disallow: with nothing after it. A file that never names an agent leaves it in a third state, and a catch-all added later blocks it without anyone deciding to.

Sites that changed their mind

last 90 days
  1. 7 October 2026 (2026-10-07)
    • ap.orgBytespider was blocked between 15 September 2026 and 7 October 2026.
  2. 5 October 2026 (2026-10-05)
    • allmusic.comBytespider was unblocked between 21 September 2026 and 5 October 2026.
    • thelancet.comBytespider was unblocked between 15 September 2026 and 5 October 2026.
  3. 3 October 2026 (2026-10-03)
    • kniga-online.comBytespider was unblocked between 15 September 2026 and 3 October 2026.
    • aflamk1.netBytespider was unblocked between 15 September 2026 and 3 October 2026.
  4. 2 October 2026 (2026-10-02)
    • dorama.landBytespider was unblocked between 15 September 2026 and 2 October 2026.
  5. 1 October 2026 (2026-10-01)
    • depositphotos.comBytespider was blocked between 17 September 2026 and 1 October 2026.
    • figma.comBytespider was blocked on 1 October 2026.
  6. 30 September 2026 (2026-09-30)
    • gemidos.tvBytespider was unblocked between 15 September 2026 and 30 September 2026.
    • getmodsapk.comBytespider was unblocked between 15 September 2026 and 30 September 2026.
    • apuestasroyal.comBytespider was unblocked between 15 September 2026 and 30 September 2026.
    • porn-comic.comBytespider was unblocked between 15 September 2026 and 30 September 2026.
    • bigbadtoystore.comBytespider was unblocked between 21 August 2026 and 30 September 2026.
    • vid30s.comBytespider was unblocked between 15 September 2026 and 30 September 2026.
    • distintas.netBytespider was unblocked between 15 September 2026 and 30 September 2026.
    • romeo.comBytespider was unblocked between 15 September 2026 and 30 September 2026.
    • insidehighered.comBytespider was unblocked between 15 September 2026 and 30 September 2026.
    • livecamrips.toBytespider was unblocked between 15 September 2026 and 30 September 2026.
  7. 29 September 2026 (2026-09-29)
    • emojicombos.comBytespider was unblocked between 15 September 2026 and 29 September 2026.
    • obqj2.comBytespider was unblocked between 15 September 2026 and 29 September 2026.

The other seven

GPTBotOpenAI

OpenAI’s crawler for gathering web content used to train its models.

OAI-SearchBotOpenAI

The crawler behind ChatGPT’s search surface: it fetches pages so they can be cited in answers.

ClaudeBotAnthropic

Anthropic’s crawler, used to gather web content for Claude.

PerplexityBotPerplexity

Perplexity’s crawler, which indexes pages so they can be cited in its answers.

Google-ExtendedGoogle

Not a crawler at all. It is a control token: Googlebot still fetches the page, and this decides whether the content may be used for Gemini and for grounding AI Overviews.

Applebot-ExtendedApple

The same shape as Google-Extended: a control token, not a crawler. Applebot fetches the page either way; this decides whether the content may train Apple Intelligence.

CCBotCommon Crawl

The crawler behind Common Crawl, a free public archive of the web that most open training corpora are built from.

How this page is made

The explanation above is written by a person. The numbers are computed from Domduck’s own daily reads of each domain’s robots.txt, parsed to RFC 9309 rather than searched for a string: a rule binds to the nearest preceding group of user-agent lines, the longest matching pattern wins, and Allow breaks a tie. No prefix matching is done, so a group naming Google is not treated as a policy about Google-Extended. What the collectors have read.