Moonshot AI · Search index
Kimi-SearchBot
Relevance scoring at fetch time is what separates this agent from an archiver: Moonshot says it "analyzes pages for relevance and builds the search index" that Kimi's search features query, so its output is a ranked corpus consulted when someone asks a question, not a static pile of text. Where the user-triggered agent retrieves the one URL a person named, this is the one deciding which URLs are candidates in the first place. The per-agent verification implied by its dedicated `kimi-searchbot.json` does not survive contact with the file: it is byte-for-byte what the other two URLs serve, so nothing at the network layer distinguishes an index sweep from a training sweep or an on-demand read. Worth noting too that four `/32` entries is the entire published fleet, no IPv6 at all: a small footprint for a consumer-scale assistant, suggesting an egress tier rather than a complete inventory.
- Operated by
- Moonshot AI
- Purpose
- Search index
- robots.txt token
- Kimi-SearchBot
- Verification
- IP verified
Crawls continuously to build and refresh an AI search index.
How to verify Kimi-SearchBot
Moonshot AI publishes the IP ranges Kimi-SearchBot crawls from, and we fetch that list on a schedule. A request claiming to be Kimi-SearchBot can therefore be checked against 4 published ranges: one that does not match is not this crawler.
Published ranges last confirmed by us on .
User agent
Moonshot AI publishes this user agent for Kimi-SearchBot. Match on the Kimi-SearchBot product token rather than the whole string: vendors revise the surrounding version and URL fragments without notice.
Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; Kimi-SearchBot/1.0; +https://www.kimi.com/policies/kimi-crawlersrobots.txt for Kimi-SearchBot
Yes, unqualified: no user-triggered escape clause is attached to this token. Moonshot documents four directive patterns for its crawlers: a full block, path-scoped blocks, `Crawl-delay: <seconds>` ("sets the minimum interval between successive requests from the same crawler"), and per-token stanzas giving different Kimi bots different rules. It also pins the scope: "Apply robots.txt per subdomain — each subdomain is treated as a separate origin."
Block
User-agent: Kimi-SearchBot
Disallow: /Allow
User-agent: Kimi-SearchBot
Allow: /robots.txt is a request, not an enforcement mechanism. It is honoured by convention, and a crawler that ignores it is stopped at your edge, not in a text file.
What blocking Kimi-SearchBot costs you
Moonshot commits to a definite outcome here where it hedges elsewhere: "Disallowing this bot will prevent your site from appearing in Kimi search results" (note "will", against the "may" it uses for its user-triggered agent). Concretely, you lose eligibility to be surfaced and cited in Kimi's answers, which for a site addressing Chinese-speaking readers is a live discovery channel and for a purely Western-market site is close to no loss at all. If crawl load rather than visibility is the real complaint, `Crawl-delay` is the documented middle path and costs no index presence. Moonshot also argues against firewalling its addresses as an alternative, because blocking the IPs can stop it reading your latest robots.txt: the quickest way to freeze a crawler on a policy you have since changed.