How-to · 3 steps
How to research TikTok Shop
with Monid.
Give this to your agent
$set up https://monid.ai/SKILL.md, then tell me what is selling on TikTok Shop and show the cost of each step
and let it take it from there.
#JobWhat came backCost
1The category15 a page, in no order$0.0015 ✓
2Find the productthe id, price and units$0.0015 ✓
3No region set111 bytes, "does not exist"$0.0015 ✓
4With region=US50 KB, and 6 left in stock$0.0015 ✓
5The creator call"no login", billed anyway$0.0015 ✓
6The one that works18 creators, 26.3M plays$0.0015 ✓
7Reviews20 rows, 12 with no text$0.0015 ✓
8Priced elsewhere$16.99 against $25.00$0.00015 ✓
9The dead route404, twice, listed stable$0.00 ✓
10Total26 billed runs$0.0795 ✓
two products · the platform was dearest on one, cheapest on the otheragent running$0.0795
Real run, 2026-09-25. A public listing, public prices.
Step 1
Set up Monid.
One line. It installs the CLI and asks for an API key from app.monid.ai. New accounts start with $1.00.
Say this to your agent
>set up https://monid.ai/SKILL.md
Step 2
Pick what you need. Say it in one sentence.
Click a job. Paste the sentence, fill in the brackets.
What is selling in a category
Say this to your agent
>get the tiktok shop products for category <id> with tikhub on monid several times, pool the results and de-duplicate, then sort by units sold yourself. there is no bestseller endpoint here and the feed is not just unordered, it is reshuffled per call: two identical requests came back with no products in common
What the agent runsdone
tikhub /shop/web/fetch_products_by_category_id$0.0015
tikhub /shop/web/fetch_products_category_list$0.0015
mrscraper /tiktok/catalog$0.036
mrscraper /tiktok/search$0.02
What came backone skincare bestseller · 2026-09-25
Units sold in one category feedlifetime units, as the feed reports them
there is no rankingnothing in this aisle returns a sales rank, and in our feed the top seller by units came back last on the page
worse than unorderedthree calls to the same category, two of them byte-identical requests, returned zero products in common; so pool several calls rather than paging one, because there is no stable sequence to page through
two documented things that are not truethe list arrives under productList rather than the documented products, and a page is fifteen rows rather than the documented twenty, so the documented offset of twenty skips five rows every time
what a month costs$0.32 to refresh a whole category's prices and unit counts every day for thirty days, at seven calls a day
listed stable, mostly notthe other provider's search endpoint answered 404 twice, and the category endpoint itself answered 400 on seven of eleven attempts while discover marked it stable; every one of those billed $0.0000
$0.001515 products a page
One listing, in depth
Say this to your agent
>get the tiktok shop detail for product <id> with tikhub on monid and set region to US explicitly. without it the call defaults to another market, returns product not exist, and bills you for it
What the agent runsdone
tikhub /shop/web/fetch_product_detail_v3$0.0015
tikhub /shop/web/fetch_search_products_list_v2$0.0015
mrscraper /tiktok/product$0.012
tikhub /shop/web/fetch_product_id_by_share_link$0.0015
What came backone skincare bestseller · 2026-09-25
what arrivesa price interval across the SKUs, every SKU's live stock, the platform's own definition of its sold count, the category breadcrumb and the seller block
the number that dates the runsix units in stock on the 100ml bottle at the moment of the call, on the listing eighteen creators were driving 26 million views at
the region defaultthis endpoint defaults to one foreign market and the older sibling defaults to a different one, so the same omission fails loudly here for a full charge and passes quietly there with a plausible answer
what it does not carryreviews: the review blocks come back empty on this call and need the dedicated one
$0.001550 KB, one call
Which creators are on commission
Say this to your agent
>get the version one product detail for <id> with tikhub on monid with an empty seller_id and region set to US, and read the related videos out of it with their play counts and their commission labels, sorting by plays yourself. use it only for the videos: on some products it returns nothing else at all. do not use the endpoint named for creator videos: it needs a logged-in cookie and an item id you do not have, and bills to tell you so
What the agent runsdone
tikhub /shop/web/fetch_product_detail$0.0015
tikhub /creator/get_product_related_videos$0.0015
tikhub /shop/web/fetch_product_detail_v3$0.0015
What came backone skincare bestseller · 2026-09-25
The creators TikTok attaches to one listingplays, top five of eighteen
the totaleighteen videos, 26,299,480 combined plays and 694,238 likes, carrying the platform's own affiliate disclosure as a field
the endpoint named for itdemands a seller-centre cookie and an item id you would need already, and bills $0.0015 to answer no login
the version trapthe call labelled "V3, full data" carries none of these; the older one the docs steer you away from is the only one that does, and on one product it returned the videos and nothing else, not even the product
absence is a signalon a second product the disclosure field was missing from exactly one video, the brand's own, so a missing label means brand-owned rather than not on commission
the sort is by likesthe videos arrive ordered by likes rather than by plays, so the first row is not the most-watched one; re-sort before you quote a top creator
what you cannot getthe creator's handle, only a display name and an id, so linking to them needs another lookup; the media urls are signed and expire in about a week
$0.001518 videos, one call
The shop behind the listing
Say this to your agent
>give me the shop stats from the detail response for <id>: lifetime units, products on sale, follower count and shop rating. do not ask how long the shop has been selling: nothing in this aisle carries a join date
What the agent runsdone
tikhub /shop/web/fetch_product_detail_v3$0.0015
tikhub /shop/web/fetch_seller_products_list$0.0015
tikhub /shop/web/fetch_seller_products_list_v2$0.0015
tikhub /shop/web/fetch_shop_id_by_share_link$0.0015
What came backone skincare bestseller · 2026-09-25
what arrives free1,495,607 lifetime units, 85 products on sale, 102,740 shop reviews, 133,857 followers, 3,834 videos and a 4.6 shop rating, all inside the call you already paid for
the older call wins againthe version one seller list returns thirty of the shop's products for $0.0015 where the newer one returns ten at the same price
what does not existhow long the shop has been selling; five endpoints' shop objects carry no join date, no opening timestamp, nothing
the link that was not recognisedthe shop-id lookup rejected the exact shop link the platform's own response had just handed us, and billed for the rejection
$0.00already in the detail call
Reviews, and how far they go
Say this to your agent
>get the reviews for <id> with tikhub on monid and keep the star histogram. count how many of the rows actually have text before you summarise: most of them are star-only, and some are reviews of a different sku
What the agent runsdone
tikhub /shop/web/fetch_product_reviews_v2$0.0015
mrscraper /tiktok/product$0.012
tikhub /shop/web/fetch_product_detail_v3$0.0015
What came backone skincare bestseller · 2026-09-25
What a page of twenty reviews holdsrows out of twenty
the histogram is freethe full one-to-five breakdown comes with page one: 3,690 five-star against 110 one-star out of 4,101, which is 90% five-star, and on a second product three endpoints agreed on the total and the histogram summed to it
the text is nottwelve of twenty rows had an empty review body, so a review-mining job needs roughly two and a half times the pages you would budget
the SKU mixfive of twenty were reviews of a sample rather than the full-size bottle, and they count toward the same total and the same 4.8 rating
the count three lines awaythe detail response carries a shop-wide review count next to the product's own, ten thousand against four thousand in the same object, and quoting the wrong one inflates the product by 2.6x
what reading them all costs$0.0015 per twenty, so 4,101 reviews is 206 calls, which is $0.31 for one product
$0.001520 rows a page
Price it somewhere else before you judge it
Say this to your agent
>price <product> on google shopping with litescrape on monid and put the merchant offers next to the tiktok shop price, adding delivery to each before you rank them. tell me where it lands among the merchants that actually came back, and tell me if the big retailers are missing from that set rather than assuming they are in it
What the agent runsdone
litescrape /google/shopping-product$0.00015
litescrape /google/shopping$0.00015
apify /axesso_data/amazon-search-scraper$0.00015 / result
tikhub /shop/web/fetch_search_products_list_v2$0.0015
What came backone skincare bestseller · 2026-09-25
One essence, one creatine, same questionUS dollars, where the platform landed
it goes both wayson the essence the platform sat at list price, tying for the dearest of fifteen merchants; on a supplement it was the cheapest offer in the whole cluster, $19.99 against a $35.50 cheapest-delivered elsewhere
the merchants may not be thereon the supplement the cluster held no supermarket, no Amazon, no big-box retailer at all: just resellers, the platform itself and three small nutrition shops, so the comparison you meant to make was not available
add the deliverythe cheapest sticker on the supplement was $25.51 and the cheapest landed price was $35.50, so ranking on the sticker alone picks the wrong merchant
before you quote any of iton one product three endpoints gave the same sku id $23.39, $42.89 and $19.99 inside four minutes, a 2.15x spread; take the price from the endpoint you will keep using and say which it was
$0.000315 merchants, two calls
Step 3
Take the cheapest route with Monid.
Everything here is a seventh of a cent. The call worth planning around is the one that prices the same item somewhere else, because it went both ways.
Find the listing · $0.0015
Search products$0.0015keyword to product id
Category feed$0.001515 a page, unsorted
Worth going deeper? · cheapest first
Price it elsewhere$0.0003
Detail, region US$0.0015
Creators on it$0.0015
The whole catalog$0.036
Reviews and stock
Star histogram$0.0015
Units left$0.00already in the detail call
Back to you
rowpriced bycost
1Price it elsewhere$0.0033
2Detail, region US$0.0048
3not listed$0.0015
4Price it elsewhere$0.0033
5Creators on it$0.0063
5 of 5 rows$0.0192
Say this to your agent
>tell me whether these tiktok shop listings are actually a deal, cheapest source first, and show the cost each: <paste product links>
“price it elsewhere first”one product sat at list price and a supermarket beat it by 32%; another was the cheapest offer anywhere
“set region explicitly”one endpoint defaults to another market, answers product not exist, and bills
“pool several feed calls”two identical requests to the same category returned no products in common
“name the endpoint you quoted”three of them priced the same sku at $23.39, $42.89 and $19.99 inside four minutes
One key. 1,700+ tools.
Fifteen endpoints on the shop, and one call next door that prices it elsewhere.
A category's whole feedtikhub · $0.0015 / call
Search by keywordtikhub · $0.0015 / call
Price, stock and the shoptikhub · $0.0015 / call
The creators on commissiontikhub · $0.0015 / call
Reviews and the star histogramtikhub · $0.0015 / call
Everything one shop sellstikhub · $0.0015 / call
A hundred in one callmrscraper · $0.036 / call
The same item elsewherelitescrape · $0.00015 / call
Amazon, by keywordapify · $0.00015 / result
+ 1,700 more across 55 providers
Browse the catalog →Read more
View all →
The best social media scraping API in 2026
The other half of the platform: what is being posted, not what is being sold.
Claude for Amazon reviews
The same product on the retailer it is 26% cheaper at, and what buyers said there.
The Apify Instagram scraper: which actor
Once you know which creators take commission, reading their other posts is the next call.