WorkerKitAI workers for Brave
Ready-made AI workers that look things up with Brave: current answers from an independent search index, for workers that need the web without leaning on someone else's results.
- Link Opportunity Finder
Finds the sites that link to your competitors but not to you, ranks them, and builds your prospect list.
What is an AI worker for Brave?
An AI worker using Brave is an AI agent that can consult the live web mid job, through a search connection backed by its own independent index rather than one resold from another engine. It fits the same shape as any other lookup step: the worker is doing something, needs a current fact, checks, and carries on with the answer in hand.
What these workers do
Check a fact mid task
Confirms a detail the job depends on, such as whether a company still trades under the name your records hold, before acting on it.
Gather background for a first contact
Reads what is publicly available about an organisation and returns a short, sourced picture of what they do.
Watch for public changes
Looks on a schedule for announcements about the organisations you track and reports what is genuinely new.
Resolve unfamiliar terminology
Looks up the industry terms in an inbound message so the reply is written by someone who understood the question.
How an AI worker runs in Brave
An index of its own
Lookups without a tracking trail
Ordinary questions, answered currently
Brave questions, answered.
How is this different from other search connections?
It runs on an independently built index rather than reselling another engine's results, so it can surface pages a different connection ranks differently or misses. When the job is to find out what is out there, that independence is the point.
Is my search activity tracked?
This connection is built around not profiling the searcher, which is a meaningful difference when a worker's queries reveal what you are working on. The questions still leave your systems as questions, so the usual advice applies: send lookups about the outside world, not about your records.
Can a worker use it to reach any website?
It searches and reads public results, in the way a person opening a few tabs does. It is not a route into systems that require a login, and it grants no access to anything inside your own connected tools, so adding search to a worker widens what it knows about the world and not what it can reach in your business.
How current are the results?
They reflect the index at the time the worker runs, so recent public changes are generally visible. For anything where the exact hour matters, the honest answer is that search is a good first check and the source itself is the confirmation.
What AI workers can do in Brave
The tools a worker can be granted in Brave once you connect it. A kit grants only the ones its job needs, and you can narrow them further on the worker.
Brave Web Search
Search the public web with Brave and get ranked results (title, url, snippet, and — where visible — content, score, published date, author, source). Set `includeContent` to pull page text inline where supported. Pass `freshness` to pin a recency window — without it the engine applies no time bound. Read the `accessInfo` field: results are filtered by the connection's egress (domain/URL allow-block) rules, so a short list may be scoped rather than exhaustive.
Brave News Search
Search recent news across the public web with Brave, returning ranked news results (same result shape as brave_search). Pass `freshness` to pin the window for dated questions (e.g. pd for yesterday's news) — without it the engine's default recency is NOT a specific window. Read `accessInfo` for any egress filtering applied.