You may be asked to provide your CreatorDB API key the first time you run a query with a skill. This is required for authentication when Claude Code calls the CreatorDB API.
Check the CreatorDB API documentation for details on the available endpoints, request parameters, and response formats to better understand how to structure your queries and interpret the results.
Check the API usage and pricing details to understand the cost implications of your queries and manage your credits effectively. For details, see API Credit Usage.
Once you have installed the CreatorDB skills, you can start using them to call the CreatorDB API through Claude Code. The CreatorDB skills are designed to be user-friendly and interactive. When you run a query, follow the on-screen instructions to provide any additional information if needed. Claude Code will then use the appropriate skill to call the CreatorDB API and retrieve the requested data. Here are some examples of how to use the skills:
Responses below are illustrative reconstructions, not verbatim terminal output — exact wording, table layout, and credit numbers will vary across runs.
To check your CreatorDB account balance, you can use the account skill. For example, you can prompt Claude Code with the following query:
Prompt example
Response example
What is my current CreatorDB account balance and API usage?
Claude Code will then use the account skill to call the CreatorDB API and retrieve your account information.
Here’s your CreatorDB account snapshot.Balance
Credits remaining: 989,923
This /usage call cost 0 credits (it’s the getApiUsage op, which is free for you).
Activity (last 7 days)
Date
Requests
Credits
2026-04-27
1
10
2026-04-28
270
2,465
2026-04-29
753
5,177
2026-04-30
326
2,074
2026-05-04
1
0
Total
1,351
9,726
Days 5/1–5/3 had no activity (omitted because the API doesn’t return zero-usage days).Where credits are goingThree contact-and-audience endpoints account for 74% of spend:
To search for a specific creator, you can use the creator-search skill. For example, to find information about the YouTube creator “PewDiePie”, you can prompt Claude Code with the following query:
Prompt example
Response example
Look up PewDiePie's profile details and related accounts.
Claude Code will then use the creator-search skill to call the CreatorDB API.
PewDiePie — Profile
Field
Value
Display name
PewDiePie
Handle
@pewdiepie
Channel ID
UC-lHJZR3Gqxm24_Vd_AJ5Yw
Verified
✅
Subscribers
110,000,000
Total uploads
4,661
Country (per CDB)
JPN (he relocated to Japan)
Main language
English
Has sponsors
Yes
Members-only content
No
Last published
2026-04-23
DB last updated
2026-04-29 (fresh)
Bio
”I make videos.”
Content profile
Category mix: Gaming 51.5% · Entertainment 48.3% · People & Blogs 0.2%
Topics: social-media comedy, variety gaming, vlogs, memes
Subscriber growth (7d/30d/90d): all flat at 0 — channel has plateaued at the 110M ceiling
Recent hashtags hint at sponsors: #ad, #partner, #shopad, #clashofclans, #clashfornature, #gfuel, #honor20pro — Clash of Clans, G FUEL, Honor (Huawei sub-brand) all visible
Other socials
Instagram: @pewdiepie
TikTok: @pewdiepie
Twitch: pewdiepie
Estimated rate card (CreatorDB model — benchmark, not a quote)
Format
CPM range (low/raw/high)
Per-video cost (low/raw/high)
Long-form video
7/10 / $12
43,844/62,635 / $81,425
Shorts
28/40 / $52
0/0 / $0 (API returned 0 — likely no Shorts pricing modeled)
Top 15 Related Creators (of 323 returned)
The full relatedCreators list contains 323 channel IDs, ranked by CreatorDB’s similarity model. Here’s the top 15 with names attached:
#
Creator
Handle
Subs
Country
1
Markiplier
@markiplier
38.6M
USA
2
penguinz0 (Charlie/MoistCr1TiKaL)
@penguinz0
18.0M
—
3
Veritasium
@veritasium
20.7M
USA
4
Doctor Mike
@doctormike
14.8M
USA
5
The Daily Show
@thedailyshow
13.8M
USA
6
LastWeekTonight
@lastweektonight
10.1M
—
7
jacksepticeye
@jacksepticeye
31.2M
IRL
8
Jimmy Kimmel Live
@jimmykimmellive
22.0M
USA
9
MrBeast Gaming
@mrbeastgaming
56.0M
USA
10
MrBeast
@mrbeast
480.0M
USA
11
CinnamonToastKen
@cinnamontoastken
4.8M
USA
12
Ryan Trahan
@ryan
22.9M
USA
13
Chat History
@chathistory
1.0M
USA
14
Technoblade
@technoblade
22.1M
USA
15
The Spiffing Brit
@thespiffingbrit
4.4M
GBR
Read
Top of the related list = closest co-watch / similar-audience peers. The first three (Markiplier, penguinz0, Veritasium) reflect PewDiePie’s actual audience overlap — gaming-comedy + science creators his viewers
also subscribe to.
MrBeast and MrBeast Gaming sitting at #9–10 makes sense given collab history and audience overlap.
Late-night TV brands (Daily Show, Last Week Tonight, Jimmy Kimmel) appearing high in the related list is interesting — likely reflects English-speaking general-entertainment co-viewing, not direct topical
similarity.
Technoblade (#14) — note this is a memorial channel; the creator passed in 2022 but the channel remains active.
The full 323-creator list is available if you want the long tail (would cost ~616 cr to enrich them all to names; happy to enrich any specific batch you point at).
Credits used this turn: 32 (1 PewDiePie profile + 15 related profiles + 1 prior balance probe was already counted). Remaining: ~991,777.
Example 3: Search for creators that meet specific conditions
To search for creators based on specific criteria, you can use the creator-search skill. For example, to find Instagram creators in the US “gaming” category with more than 1 million followers and high engagement rates, you can prompt Claude Code with the following query:
Prompt example
Response example
Find me the top 10 US gaming Instagrammers with at least 1M followers and high engagement rates
Claude Code will then use the creator-search skill to call the CreatorDB API.
Top 10 — US Gaming Instagram, 1M+ followers, ranked by Reels engagement rate
Here’s the ranked top 10. Worth flagging some quality caveats below.
#
Handle
Display
IG Category
Followers
ReelsER
1
@nihaachuu
Niki
Video creator
1,414,162
6.98%
2
@nrggram
NRG
E-sports team
1,590,020
4.62%
3
@thepointerbrothers
the pointer brothers
Digital creator
2,510,184
4.60%
4
@shroud
Michael Grzesiek
Public figure
1,258,981
4.20%
5
@thesketchreal
Sketch
Digital creator
1,716,511
4.16%
6
@nadeshot
Matthew Haag
Digital creator
1,161,500
4.06%
7
@annashumatee
anna
Gamer
1,465,704
3.03%
8
@ronnie2k
Ronnie Singh
Public figure
1,098,515
3.01%
9
@lolesports
LoL Esports
ESPORTS_LEAGUE
1,838,146
2.69%
10
@g2esports
G2 Esports
ESPORTS_TEAM
1,378,395
2.57%
Read
Confirmed core-gaming creators (8 of 10): shroud, Sketch, Nadeshot, ronnie2k (NBA 2K), and the
four esports orgs (NRG, LoL Esports, G2, plus the “Gamer”-categorized annashumatee). These are
unambiguously gaming.
Verify before outreach (2 of 10):
#1 @nihaachuu (6.98% ER) — IG category is “Video creator”, not “Gamer”. She showed up in the gaming-niches search but her primary content is modeling/lifestyle, not gameplay. ER is genuinely high but she may not be a fit if you specifically want gaming-content creators.
#3 @thepointerbrothers — also “Digital creator”. They make NBA / sports-comedy content; appeared here because they used #gameday / #gaming hashtags. Almost certainly not a gaming creator.
Industry benchmarks: <0.5% is typical for mega-IG accounts, 1–3% is healthy mid-tier, >5% is exceptional. The whole list lands in the healthy-to-exceptional range, which is plausible for gaming-niche IG (smaller, more loyal audiences).
Methodology caveat
Instagram’s gaming discovery in CDB is hashtag-driven and noisy:
The id_gaming_All niche pulled 23K creators including HOT 97, Demi Lovato, Sanrio, Matthew Lillard.
NLS for the same query surfaced Zendaya, Cardi B, Rihanna alongside real gaming creators.
The IG category field (e.g., “ESPORTS_TEAM”, “Gamer”) is more reliable than niches but only ~5–10% of creators have a gaming-specific value.
So this list is the result of: niche-search candidates → manual filter to gaming-clear handles → enrich with /instagram/performance for actual ER → rank. If you need a strictly-vetted list, drop @nihaachuu and @thepointerbrothers and pull the next two candidates (@dakotaz at 1.49% ER and a couple from a wider search).
To create an outreach list of potential creator partners for a campaign, you can use a combination of the creator-search and creator-enrichment skills. For example, to find 50 beauty TikTokers with more than 500,000 subscribers in the US with high engagement rates and retrieve their contact information, you can prompt Claude Code with the following query:
Prompt example
Response example
Find the top 50 beauty TikTok creators with more than 500,000 subscribers in the US with high engagement rates and retrieve their contact information.
Claude Code will then use the creator-search and creator-enrichment skills to call the CreatorDB API.
Top 20 — US Beauty TikTok, 500K+ followers, ranked by engagement rate
Rows 21–50 are saved in /{YOUR WORKING DIRECTORY}/{FILENAME}.csv with full bios, niches, and verification flags. Notable agency contacts in the long tail: kensingtongrey, dulcedo, undercurrent.net, viralnationtalent, fixated, lucky8collective, thedigitalbrandarchitects.
Two important caveats
The ER values are TikTok-formula (likes + comments + shares) / followers, which runs noticeably higher than YouTube/Instagram ER. #1 @cannolifamily at 97.05% is almost certainly a recent-window artifact (one viral video pulling the rolling average up); double-check before reporting it as a sustained metric. #2 @michaelahallmark1 at 34% is also extreme — verify content fit (her bio mentions “MakeJesusKnown” hashtags, may be more lifestyle/faith than beauty).
Niche-tag noise: The TikTok beauty niches are hashtag-driven, and several handles in the list don’t clearly skew beauty:
@still.bad.decisions (#20) — bio says “TheCarGirl BackUp Cars, corgs, and complaining”
@dunknatachai (#34, 3.2M) — Thai actor (Disney/GMMTV credits)
@sway_bhatia1 (#31) — actress
@christwodafur (#38, 3.1M) — Star Wars cosplay
Some have very generic bios suggesting general lifestyle content rather than beauty
The strongly-beauty-confirmed handles are the ones with agency emails (kensingtongrey, dulcedo, etc.) and bio mentions of makeup/skincare. If you want a strictly beauty-only roster, filter the CSV to rows with beauty-keyword bios or pull /tiktok/sponsorship to confirm beauty-brand work history.
Summary
Metric
Value
Rows processed
50 / 50
Failed rows
0
Rows with ≥1 email
37 (74%)
Bio-email hits (free)
27
/tiktok/contact calls
23
Rows with no email
13
Credits
Step
Cost
/tiktok/search × 1
1 cr
/tiktok/profile × 50
100 cr
/tiktok/contact × 23
345 cr
/tiktok/performance × 50
100 cr
Total
546 cr
Bio-first contact approach saved ~405 cr vs. blindly calling /contact on all 50. Remaining: ~987,650.
Examples 5a-5c each show a different cut of the brand-sponsorship skill: known brand → roster (5), vertical → brands (6), brand → spend estimate (7).
The following examples show how to use the brand-sponsorship skill to analyze a brand’s sponsored creator roster, identify which brands are active in a specific vertical, and estimate a brand’s monthly spend on creator partnerships.
To verify a brand’s sponsored creator roster, you can use the brand-sponsorship skill. For example, to find out which creators are sponsored by Adidas, you can prompt Claude Code with the following query:
Prompt example
Response example
List all creators sponsored by Adidas.
Claude Code will then use the brand-sponsorship skill to call the CreatorDB API.
Adidas — All Sponsored Creators
Total: 432 creators — 405 YouTube + 27 Instagram. Saved to /{YOUR WORKING DIRECTORY}/{FILENAME}.csv.
Headline numbers
Platform
Creators
Notes
YouTube
405
Long tail of football/sports niche channels
Instagram
27
Mostly athlete partnerships, much sparser index
YouTube — geographic spread
Top 10 countries by creator count: USA 69 · ESP 25 · JPN 23 · GBR 17 · FRA 17 · THA 17 · POL 16 · KOR 16 · BRA 15 · DEU 15True global program — not US-centric. Spain, Japan, and Thailand are notably high (football market + Asian apparel push).
Most-active YouTube partners (by # placements with Adidas)
Creator
Country
Subs
Placements
Last
Fotbollsmorgon
SWE
40K
25
2026-04-04
Canal Camisas e Chuteiras
BRA
187K
19
2026-03-02
John Cincola Pickleball
USA
80K
19
2026-02-26
FC AFKICKEN
NLD
50K
18
2025-11-07
Soccer Reviews For You
USA
836K
16
2025-10-01
futbolmania
ESP
60K
15
2025-10-02
Best Buy Soccer
USA
<10K
14
2026-01-15
골챔 (Golchamp)
KOR
13K
13
2025-12-03
SUPERGAANDE
NLD
887K
11
2025-05-28
Futbolistas
MEX
871K
11
2025-08-20
Pattern: Adidas leans heavily on mid-tail soccer/football review channels for repeat partnerships (boots, kits, gear reviews) rather than mega-creators. Volume of placements doesn’t track follower count — Fotbollsmorgon has 25 placements at just 40K subs, while 12.9M-sub J Jai Pan has only 1.
Largest YouTube creators (by reach)
12.9M (J Jai Pan, THA) · 8.57M (loveliveserve, USA) · 6.91M (Complex, USA) · 5.40M (Unisport, 7 placements — also a heavy partner) · 5.40M (Blowek, POL) · 2.89M (DAZN Football) · 2.32M (adidas’ own channel — interesting that it shows up as a “sponsored creator” of itself, treat as a self-reference noise row).
Instagram — much sparser
Only 27 IG creators indexed. Top by reach: N’Golo Kanté (16.5M, 2021), Trinity Rodman (750K, 2026-01), Melanie Leupolz (290K, 2024). Most have just 1 placement and many entries are very small accounts (<50K). The IG sponsored-content index for Adidas appears thin compared to YouTube.
Caveats
adidas channel itself appears in the YouTube list (USA, 2.32M subs, 1 placement). That’s the brand’s own channel — likely a self-tag from a user-generated content collab. Filter it out for clean creator-side analysis.
CreatorDB sponsorship detection is heuristic — undisclosed paid partnerships and creators using non-standard tags are missed. Real Adidas roster is likely larger.
The 27 Instagram results is a suspiciously low count for a brand of Adidas’s scale — IG sponsorship indexing in CDB is generally less complete than YouTube’s. If IG is critical, treat this list as partial.
Credits
Step
Credits
/sponsor/information
2
/sponsor/creators (YouTube ×5 pages)
125
/sponsor/creators (Instagram ×1 page)
25
Total
152
Remaining: ~990,189.CSV columns: platform, creatorId, displayName, country, followers, sponsoredCount, lastSponsoredDate, avgRecentSponsoredEngagementRate, topics, niches. Open from your working directory with Numbers or Sheets.
To retrieve data on brands sponsoring creators, you can use the brand-sponsorship skill. For example, to find the top 10 brands sponsoring US gaming YouTubers, you can prompt Claude Code with the following query:
Prompt example
Response example
Find the top 10 brands sponsoring US gaming YouTubers.
Claude Code will then use the brand-sponsorship skill to call the CreatorDB API.
Top 10 Brands Sponsoring US Gaming YouTubers
Based on a sample of 50 US gaming YouTubers (1M+ subs, English-speaking, active), aggregating who sponsors how many distinct creators.
CoryxKenshin, VanossGaming, theRadBrad, The Game Theorists
3
Marvel Rivals
Game (NetEase)
4
5
H2ODelirious, VanossGaming, ItsFunneh, The Game Theorists
4
Intel
Hardware
3
6
MrBeast Gaming, SypherPK, Caylus
5
AFK Journey
Game (Farlight)
3
4
VanossGaming, The Game Theorists, Caylus
6
Crystal of Atlan
Game (NVS/NetEase)
3
3
VanossGaming, The Game Theorists, Caylus
7
Sony
Hardware/Publisher
3
3
VanossGaming, theRadBrad, SypherPK
8
War Thunder
Game (Gaijin)
3
3
KreekCraft, VuxVux, Caylus
9
Chime
Fintech
2
8
UnspeakableReacts, The Game Theorists
10
NordVPN
Cybersecurity
2
6
Markiplier, The Game Theorists
Read
Game publishers and individual game titles dominate — 7 of the top 10 are gaming companies promoting their own games. Epic Games leads on roster reach (Fortnite-related promos especially), Ubisoft leads on placement volume (15 placements across 4 creators = repeat partnerships).
Mobile gacha / F2P MMOs are aggressive sponsors — Marvel Rivals, AFK Journey, Crystal of Atlan, War Thunder all rely heavily on creator marketing for user acquisition. NetEase appears under multiple titles (Marvel Rivals + Crystal of Atlan).
Non-gaming partners: Chime (banking) and NordVPN (VPN) — the two evergreen “every creator” sponsors. Chime’s 8 placements across just 2 creators shows it’s a high-frequency repeat partner where it does land.
Hardware tier: Intel (3) and Alienware (2, just outside top 10) — the consistent PC-hardware pair.
Just outside top 10: Alienware, 2K, Lurkit (a creator-marketing platform itself), GeoGuessr, my.games — all at 2 creators each.
Caveats
Sample of 50 creators, sorted by subscribers descending — top-of-funnel skew. Smaller channels in the long tail may show different sponsor patterns.
CreatorDB sponsorship detection is heuristic (disclosure tags, hashtag conventions). Undisclosed paid placements get missed; the actual ranking could shift slightly.
/youtube/sponsorship returns up to ~20 most recent sponsored uploads per creator, not lifetime — so very old partnerships won’t appear.
Tie-breaking near the cut: brands ranked #8–10 all have only 2–3 creators in the sample. A larger sample (e.g., 200 creators) would give more robust ranks below the top 5.
Refinements available
Expand the sample to 200 creators (~1,000 cr more) for tighter ranking below the top 5
Filter to a specific gaming sub-niche (e.g., FPS, Roblox, Minecraft) using the niches filter
Cross-reference with /sponsor/summary for the top brands to see their estimated 30-day spend on YouTube overall
Get the audience demographics of each top brand’s sponsored pool with /sponsor/audience
Credits
Credits used this turn: ~251 (1 search + 50 sponsorship). Remaining: ~991,558.
To retrieve data on a brand’s spend on sponsoring activities, you can use the brand-sponsorship skill. For example, to find Nvidia’s latest sponsoring activities, you can prompt Claude Code with the following query:
Prompt example
Response example
Estimate Nvidia's influencer marketing spend over the last 30 days
Claude Code will then use the brand-sponsorship skill to call the CreatorDB API.
Nvidia — estimated influencer spend, last 30 days
~$24,448 USD (YouTube only; CreatorDB model estimate)
Context
Metric
Value
Sponsored content (30d)
22 pieces
Views on that content (30d)
1,474,969
Estimated CPM (30d)
$16.58
Estimated CPE (30d)
$0.51
Avg engagement rate
3.93%
7d spend / 90d spend
0/249,141
30d view growth vs prior 30d
−5.3%
30d engagement growth
−2.3%
Read
The 90-day total is ~249K,sothelast30days(24K) is well below the trailing pace — Nvidia’s sponsored YouTube cadence has cooled meaningfully. The $0 in the last 7 days reinforces that (no detected sponsored placements in the most recent week).
Engagement and views are both negative on the 30-day growth window — fewer placements and softer per-piece performance.
Instagram is excluded. Nvidia has 0 sponsored creators indexed on Instagram, and /sponsor/summary doesn’t model IG spend regardless — so this number is effectively the full picture for the platforms CreatorDB tracks (YouTube + IG), but it does not include TikTok or other channels.
Caveats
These are modeled estimates, not booked spend — useful for benchmarking and trend, not for citing as Nvidia’s actual budget.
Engagement rate here uses (likes + comments) / views, different from creator-side engagement formulas — don’t compare directly to creator search filters.
Example 6: Transcribe a video for content analysis
To transcribe a video for content analysis, you can use the content-analysis skill. For example, to transcribe a creator’s video to find out if they mention a specific brand, you can use the following prompt:
Prompt example
Response example
Transcribe MrBeast's most recent YouTube video and find any brand mentions in it.
Claude Code will then use the content-analysis skill to call the CreatorDB API.
MrBeast's Most Recent Transcribable Videos — Brand Mentions
The transcript availability situationOf MrBeast’s most recent uploads:
Date
ID
Type
Title
Subtitle status
2026-04-29
EdT5Z16Y5WE
Short
Would Your Dog Pick $10,000 or Steak?
❌ none indexed
2026-04-28
ScjQAcbA9e4
Short
Important Announcement From The Grocery Store
❌ none
2026-04-27
hRBpDdWbUpI
Short
Can My Fiancée Lift Me?
❌ none
2026-04-24
pzxtGL9kzPs
Short
Try Not To Get Slimed
✅ auto English
2026-04-21
yy-Nt0dRhNI
Short
Winner Picks Our Honeymoon
✅ auto English
2026-04-18
zRtGL0-5rg4
Long-form
Last To Leave Grocery Store, Wins $250,000
❌ still none
The actual most-recent long-form video still hasn’t had subtitles ingested by CreatorDB.Transcripts of the two most-recent transcribable videospzxtGL9kzPs “Try Not To Get Slimed” (2026-04-24, ~30s)
“OH, GOSH. WHAT IS THAT? WOAH, WOAH, WOAH, WHAT? ARE YOU DONE? YEAH, GO AWAY. THAT’S DISGUSTING. … OH, IT FEELS SO NASTY, BRO. KEEP POURING. ALL RIGHT, HERE WE GO. … IT’S SO BUBBLY.”
“Winner picks our honeymoon. Okay, you just got to get more balls in than the other person … I want to go on an island. … Right after we get married WITHOUT MY PHONE, BABE. … I want uh to go to space. … Hey, Jeff Bezos, you know that like rocket that close your ears that looks kind of like a […] that goes to space. I would love to go.”
Brand / public-figure mentions
Mention
Context
Type
Jeff Bezos
Direct address asking for a Blue Origin ride
Founder reference
Blue Origin (implied)
“rocket that close your ears… goes to space” — unmistakable reference to New Shepard’s well-memed silhouette
Indirect product reference
That’s the entire brand surface area in these two shorts. No Feastables, no sponsor reads, no product placements.
Example 7: Submit and enrich creator list in CSV format
To submit and enrich a creator list in CSV format, you can use the creator-csv-workflow skill. For example, if you have a CSV file with a list of YouTube creator channel IDs and you want to enrich it with additional data from CreatorDB, you can follow these steps:
1
Get a list of creator account or channel IDs.
a. Obtain YouTube creator channel IDs through any of the following methods:
Call the CreatorDB API.
You can call the CreatorDB API using the creator-search skill in Claude Code to find creators based on specific criteria (e.g., niche, subscriber count, location). This will return a list of creator profiles, including their channel IDs, which you can then export to a CSV file for enrichment. For details, see Example 2 and 3 in this guide on how to use the creator-search skill effectively.
Search social media platforms, such as YouTube.
Use your own database.
Purchase a list from a data provider.
b. Ensure that the channel IDs are in a CSV format with a header row (e.g., channelId).
2
Prompt Claude Code to enrich the creator list.
Prompt Claude Code with the following query, replacing path/to/your/creators.csv with the actual path to your CSV file:
Enrich the creator list in this CSV file with additional data from CreatorDB: path/to/your/creators.csv
Claude Code will then use the creator-csv-workflow and creator-enrichment skills to call the CreatorDB API and enrich your creator list with additional data such as subscriber count, engagement metrics, audience demographics, and more. The final output should be an enriched CSV file with the new data columns added, which you can use for further analysis or outreach. For example, if your original CSV had a channelId column, the enriched CSV might include additional columns like displayName, totalSubscribers, engagementRate, audienceAgeRange, etc., depending on the data available for each creator in CreatorDB. The following is an example of the messages returned for a successful enrichment of a list of 50 gaming YouTubers:Results
Metric
Value
Rows processed
50 / 50
Failed rows
0
Rows with ≥1 email
37 (74%)
Bio-email hits (free)
11
/contact calls made
39
Total credits used
1,185
Bio-first saved ~210 cr vs always calling /contact (would’ve been 50 × 27 = 1,350).Output
