Best of · Agents

The best data APIs for AI agents one pick for each job.

The best data source for an AI agent depends on the job: LinkedIn for people and companies, Google Maps for places, Amazon for products, Indeed for jobs, Booking.com for stays, Google for the open web and Reddit for opinions. All of them answer through one account here, so the agent signs itself up once, quotes each call and pays per record.

Platforms
26
Sign-up
One call
Free trial
$0.10
Quote
Free, before each call

Updated 21 October 2026

One pick per job

Show all 10 rows
1People and companies: LinkedInProfiles, companies, people search, posts, jobsWho works wherefrom $0.001 per record
2Places: Google MapsPlace cards, search, reviewsAny business with an addressfrom $0.000225 per record
3Products: AmazonProduct cards, search, reviews, price historyWhat things cost and what buyers sayfrom $0.00045 per record
4Jobs: IndeedFull postings with salary, skills, employer signalsWho is hiringfrom $0.000045 per record
5Stays: Booking.comProperties, search, guest reviewsHotel prices for datesfrom $0.0018 per record
6The open web: GoogleWeb results, news, images, shopping, Play reviewsEverything else that is publicfrom $0.00009 per record
7Opinions: RedditPosts with score, comments, searchWhat people really thinkfrom $0.0045 per record
8Creators: InstagramProfiles, posts, hashtags, followers, similar accountsInfluencers and brandsfrom $0.00045 per record
9Contact details: ContactsWork email and mobile number by LinkedIn handleReaching a personquoted per call
10The long tail: ExploreRaw rows from sources outside the platform toolsWhen nothing above fitsquoted per call

List prices are per record and some calls carry a minimum. Quote any call first with "estimateOnly": true: the quote is free and is the most the call can cost.

Each source, in order

1. People and companies: LinkedIn

Profiles with work history, company pages, and people search by title, seniority and company. The richest source for questions about people at work. Company-page posts are not served.

2. Places: Google Maps

Search a keyword in a location, read places by URL and pull reviews. Cards carry phone, website, hours and rating.

3. Products: Amazon

Products and search on 12 marketplaces, reviews with verified-purchase flags, and a free price history.

4. Jobs: Indeed

Keyword and place search on 23 country sites; every hit is a full posting. Use LinkedIn jobs beside it for professional roles.

5. Stays: Booking.com

Search a destination for a stay and read rooms with availability and price. Without dates there is no price.

6. The open web: Google

Ranked links and snippets with search operators, plus news, images and shopping offers. Web search is billed per page of about ten results.

7. Opinions: Reddit

Search all of Reddit by keyword, read a subreddit by sort, and pull comment threads. There is no user profile card.

8. Creators: Instagram

Profiles with recent posts in one call, hashtag feeds and similar accounts. Free history tracks follower growth.

9. Contact details: Contacts

Up to ten people per call by /in/ handle, with an optional email check. Priced per call, billed per person and per part; the one source here that is not read from public pages.

10. The long tail: Explore

Search what is reachable for free, read a source's input fields and estimate, then run it. Rows come back raw and unverified, not mapped into a schema, so use the platform tools first.

Why one account matters to an agent

An agent that needs five sources should not need five sign-ups, five keys and five bills. Here it registers itself in one call, starts with $0.10 free, and reads the operations as plain text from llms.txt.

Before it spends, it can quote any call free with "estimateOnly": true and cap it with "maxUsd". Every answer says when each row was measured, what was missing and why, and what the call cost, so the agent does not invent the gaps.

The operations on this list

Questions people ask.

Can an AI agent buy data on its own?

Yes. It registers itself with one POST and starts on a $0.10 trial. A person claims the agent later with a code, only when it is time to pay.

How does an agent avoid overspending?

It spends a prepaid balance, quotes calls free with estimateOnly, and maxUsd refuses anything above a ceiling. Only a person can add money.

What data is not covered?

Anything behind a login: private profiles, inboxes and messages. Contacts (work email and phone) is the one source that is not read from public pages.

How many platforms are there?

26, from LinkedIn, Instagram and TikTok to Amazon, Blinkit, Indeed, Booking.com and Google, plus Explore for the long tail.

Try it in a minute. $0.10 free.

The playground runs any endpoint by hand: pick a platform and what you want from it, fill a short form named in plain words (or press Use sample input), see the free quote on the Run button, run it, then read the answer as a table or JSON and download it as CSV or JSON. The same call is written out as cURL, Python and Node while you type.

Try in playground

Quickstart · Every price · All platforms

Why people pick API for Agents

  • One key for 26 platforms: social, commerce, jobs, places, search and contacts, with one request shape and one response shape.
  • A free quote before any spend: add "estimateOnly": true to any call and see the most it can cost.
  • Pay per record in US dollars. No subscription, nothing expires, top-ups from $5.
  • $0.10 free to try, no card. An AI agent can sign itself up in one request; a person claims it later.
  • Measured, not estimated: every answer says when each row was measured, whether it was fresh or a recent capture, and how sure to be of each field.
  • Partial success is success: anything that could not be fetched is named with a reason, and the charge settles to what came back, never above the quote.
  • Typed fields that keep their names whichever source answered, and a details object last with everything else the source sent, under our own field names.
  • Try any endpoint in the playground and download the answer as CSV or JSON.

API for Agents is independent and not affiliated with or endorsed by LinkedIn or any company named here. LinkedIn is a trademark of LinkedIn Corporation; other names belong to their owners. Public data only: nothing behind a login.