Always the same name
Every request carries the User-Agent above: no rotation, no disguise, no borrowed browser name.
The crawler behind Nagare. It says who it is, follows your robots.txt and never tries to get around you.
Nagare keeps a dated record of public web pages for product teams: what changed, when, and what it means. NagareBot is the robot that reads those pages.
NagareBot/0.1 (+https://nagare.watch/bot)
The version number changes. The name NagareBot does not.
Someone using Nagare follows your company. NagareBot reads the public pages of your site that matter to them — pricing, product, docs, changelog, careers — on a schedule: once a day or once a week, sometimes on request.
It reads what anyone can see without an account, and nothing else. Nagare does not disclose which customers follow a site.
Every request carries the User-Agent above: no rotation, no disguise, no borrowed browser name.
NagareBot follows the rules you write for it, or those for every robot (*) when there are none. It reads your robots.txt again at least once an hour. If your robots.txt answers with a server error, it reads nothing.
When it reads your server directly: one request at a time, at least a second apart, longer if your Crawl-delay asks for it. A 429 or a server error makes it pause; it tries again a few times at most, then moves on.
It follows links within your site only. Links to other sites are noted, never followed.
It never logs in, never fills in a form and never tries to get past a paywall.
No disguise, no anti-bot bypass, no CAPTCHA solving. If you block NagareBot, it stays blocked.
NagareBot reads a site either directly, from Nagare’s servers, or through Firecrawl, a service that renders pages the way a browser does, for sites that need JavaScript. Firecrawl’s requests carry the NagareBot User-Agent too and come from Firecrawl’s servers; Firecrawl applies robots.txt for its own robot as it reads. Nagare then checks every page against your rules for NagareBot, and keeps nothing NagareBot is not allowed to read.
Everything goes through your robots.txt. A change applies within the hour.
User-agent: NagareBot
Disallow: /User-agent: NagareBot
Disallow: /private/User-agent: NagareBot
Crawl-delay: 10Seconds between two requests, when it reads your server directly.
Blocking it in robots.txt is enough.
A question, a problem, a request about pages already read: write to bot@nagare.watch.
Last updated 20 October 2026.