AI2Bot-DeepResearchEval
AI2Bot-DeepResearchEval is a bot operated by Ai2, a non-profit AI research institute that fetches single pages on demand for an AI assistant. Ai2Bot-DeepResearchEval is operated by Ai2, a non-profit AI research institute. No IP range list or reverse-DNS convention is published, so a request carrying this user agent cannot be proven genuine and a robots.txt rule alone will not stop a forgery.
AI2Bot-DeepResearchEval at a glance
| User agent | AI2Bot-DeepResearchEvalToken only. The operator has not published a full user-agent string, so match on the substring rather than an exact header. |
|---|---|
| Operator | Ai2, a non-profit AI research institute |
| Purpose | Assistant fetchers |
| Respects robots.txt | Not documented |
| Verification method | None published, user agent only No IP range file and no reverse-DNS convention have been published for this agent. The user-agent string is the only signal, and any client can send it, so treat a match as a claim rather than proof. |
| Crawl pattern | Not published |
| Robots.txt tokens | AI2Bot-DeepResearchEval |
| Operator documentation | None published |
Ai2Bot-DeepResearchEval is operated by Ai2, a non-profit AI research institute. It's used to collect and scan resources used in deep research queries performed by Ai2's o…
Allow or block AI2Bot-DeepResearchEval
Paste one of these into the robots.txt file at the root of your domain. Rules are per token, so a block on one crawler leaves every other bot untouched.
User-agent: AI2Bot-DeepResearchEval
Disallow: /Blocks every path for this crawler only.
User-agent: AI2Bot-DeepResearchEval
Allow: /Explicit allow. Useful when a wildcard rule above it would otherwise catch this crawler.
User-agent: AI2Bot-DeepResearchEval
Allow: /
Disallow: /account/
Disallow: /checkout/
Disallow: /searchEdit the Disallow paths to match your own account, checkout and search URLs.
What blocking actually costs you
Blocking AI2Bot-DeepResearchEval means a person who asks about one of your URLs gets an error instead of a summary. That is a support and reputation trade-off, not a training one.
Block every crawler in the directory at onceIs that really AI2Bot-DeepResearchEval?
A user-agent header is a string the client chooses. Scrapers copy AI2Bot-DeepResearchEval precisely because site owners allow it. No IP range file and no reverse-DNS convention have been published for this agent. The user-agent string is the only signal, and any client can send it, so treat a match as a claim rather than proof.
Paste the IP address from your access log below. You will see whether it belongs to a hosting provider, a residential proxy pool or a Tor exit, plus the ASN that announces it, which is what tells you whether the claim holds up.
Related crawlers
Identify crawlers automatically instead of by hand
Agentscan checks a request against published crawler ranges, reverse DNS and hosting data, then returns a verdict your edge can act on. One call per request, no range files to keep up to date.