Skip to content
GrokBotsHub

Open data

Developers: feeds and search API

The directory is public, and so is its data. Use the feeds and the search API in your own tools, with fair use and a link back.

Feeds

Both are free to read. Please poll them at most once an hour.

GET /api/search?q=<text> searches bots, jobs, apps and guides. It is the same search the site's search box uses. It is public and read-only, and needs no key.

curl "https://grokbotshub.com/api/search?q=inbox"

The answer is JSON in this shape:

{
  "hits": [
    { "type": "bot",   "title": "…", "sub": "by … · …", "href": "/bots/…" },
    { "type": "job",   "title": "…", "sub": "…",        "href": "/jobs/…" },
    { "type": "app",   "title": "Works with …",          "href": "/apps/…" },
    { "type": "guide", "title": "…", "sub": "…",        "href": "/learn/…" }
  ]
}
  • type is bot, job, app or guide. sub can be missing. href is a path on grokbotshub.com.
  • You get up to 8 bots, ranked by relevance, then up to 3 jobs, 3 apps and 3 guides whose name contains your text.
  • q needs at least 2 characters; a shorter one returns {"hits":[]}. Only the first 80 characters are used.
  • Responses may be cached for 60 seconds.
  • The limit is 120 requests per minute per IP address. Above it you get HTTP 429 with {"hits":[],"error":"rate_limited"}.

Fair use

  • Cache what you fetch. Send one request at a time.
  • Send a User-Agent that names your project and a way to reach you.
  • Do not rebuild the directory as a copy of the site. Link to it instead.
  • Need more than the limits allow? Write to [email protected] and tell us what you are building.

Attribution

When you show our data, credit it as "Data: GrokBotsHub" and link each bot to its page on the Hub. Trust levels, estimates and Hub test reports are our work: quote them with a link and the date they carry. Listing text belongs to each bot's creator, and share pages belong to x.ai. The terms of use apply.