Provider: AI Research's web sources

AI Research reads the open web while it looks for matches. This article covers which sources it can reach and what's free versus paid; for exact per-call pricing and how metering works, see "How is AI Research metered, and what do paid sources cost?"

What it delivers. A free Google-powered web search, scoped on demand to a specific site (Reddit, Hacker News, GitHub, LinkedIn or a company's own domain) or left unscoped for press and team pages; a free news search for announcements and funding news; a free, structured Hacker News search (story and comment authors); open job postings as a free hiring signal; a free page-reading tool that fetches the text of any URL the agent finds (a forum thread, a team page, a GitHub profile); and a free LinkedIn lookup that grounds a real profile before a match is returned. On top of that, up to five opt-in paid sources: a B2B database lookup, LinkedIn post search (plus the people engaging with a post), Reddit search, Twitter / X search and a BuiltWith technology lookup, for records, posts, profiles and tech-stack data the free sources don't otherwise surface well.

Where in ListPlus. Runs automatically during every AI Research run; every source is its own switch under Allowed sources & tools. Web search, reading pages, job postings, news, Hacker News and finding a real LinkedIn profile are free and on by default; the database lookup, LinkedIn posts, Reddit and X are paid per call and off by default until you tick them (the BuiltWith lookup is currently not available).

Inputs needed. The research brief you describe when starting a run — AI Research builds its own search queries from it.

Coverage & freshness. Whatever is currently indexed or postable on each site at the moment of the search — there's no stored copy or delay.

Credits & charge rule. Web search, news search, Hacker News search, job search, page reading and the LinkedIn lookup are always free. The database lookup, LinkedIn post search, Reddit search and Twitter / X search are billed per call, whether or not the call finds anything; see "How is AI Research metered" for current rates. The BuiltWith technology lookup is currently not available.

What the provider receives from us. Only the search query the agent builds for that one call, never your list or brief in full.

Company seat, data location, legal basis. Available on request from support@listplus.ai.

Caching. Not cached — a research run searches live each time.

Related topics: research-costs-and-sources · ai-research · provider-google-search.

FAQ#

Which web sources can AI Research search?#

Google-powered web search (which it can scope to Reddit, Hacker News, GitHub, LinkedIn or a company's own site), a news search, a free Hacker News search, job postings, a page-reading tool that fetches any URL it finds, and a LinkedIn lookup to ground real profiles. A B2B database lookup, LinkedIn post search, Reddit, Twitter / X and a BuiltWith tech-stack lookup are five further, opt-in paid sources.

Is there a dedicated GitHub source, or does it use web search?#

There's no separate GitHub tool — the free web search is scoped to github.com when the agent judges that useful, the same way it scopes to Reddit or Hacker News.

Are Reddit, X and BuiltWith free like the others?#

No — those are opt-in paid switches under Allowed sources & tools, alongside the database lookup and LinkedIn post search; every free source (web search, news, Hacker News, job postings, reading a page, finding a real LinkedIn profile) is on by default. Reddit search and Twitter / X search cost credits per call, each charged whether or not it finds something; the BuiltWith technology lookup is currently not available. See "Price list: what every action costs" for the current rates.

How does AI Research read a company's own website or team page?#

The same free page-reading tool that opens any URL the agent finds — it fetches the text of a team, about or press page once a search points to it.

Does AI Research use a stored index, or search live each time?#

Live each time. Every source reflects whatever's currently on the web or postable at the moment of the search, with no stored copy or delay.

What does a source receive when AI Research calls it?#

Only the search query the agent builds for that one call — never your full research brief or list.