Is YouTube search interest the same as the YouTube Trending board?

No. source youtube on get_growth or get_time_series returns a 0-100 keyword search-interest series. type YouTube Trending on get_top_trends returns ranked video titles with no keyword field. The September 6, 2026 pull put Minecraft at search index 8.0 for the week ending 2026-09-05 while four Minecraft titles sat on the live board. A category can dominate the board while its search index stays in single digits.

What happens if youtube and YouTube Trending are swapped?

The call fails. type youtube on get_top_trends returned HTTP 400 error invalid_request and listed 24 valid feed names. source YouTube Trending on get_growth returned HTTP 400 error invalid_source and echoed youtube trending in lowercase, with 16 valid keyword sources including python. Pass the lowercase source to growth and series tools. Pass the exact feed label to get_top_trends.

Does omitting limit change the YouTube Trending board?

Omit limit and the September 6, 2026 omitted-country snapshot returned limit 25, offset 0, count 25, as_of_ts 2026-09-06T06:01:36.749501+00:00. That default is a cap on the payload, not the full stored board. A 7d rank_change pull against the same snapshot returned a title at rank 101 with prev_rank 192, so the archive is deeper than 25 rows.

Can a new video title be used as a youtube keyword?

Yes, but read the quality flags. SaWaDiKa on source youtube printed 7D growth 2750.0% from baseline 2.0 on 2026-08-29 to 57.0 on 2026-09-05, with data_quality low and a warning that the underlying signal is over 90% zero values. The matching 12M row printed growth 999999.0 because baseline_value was 0.0. Prefer a stable artist or franchise keyword when the title string is this sparse.

How are the two calls billed?

get_growth bills one request per source plus keyword. Extra percent_growth windows in that call are included. get_time_series bills one request per source plus keyword. get_top_trends on MCP costs 1 credit per 10 requested rows, max 10 credits per call. REST bills 1 request per get_top_trends call. The Free plan is 100 requests per month.

YouTube Search vs YouTube Trending API

YouTube search interest and the YouTube Trending board are two different Trends MCP calls. Pass source youtube to get_growth or get_time_series with a keyword. Pass type YouTube Trending to get_top_trends with no keyword. On the September 6, 2026 pull, Minecraft sat at YouTube search index 8.0 for the week ending 2026-09-05, down 11.11% from 9.0, while four Minecraft titles occupied ranks 2, 8, 11, and 17 on the omitted-country YouTube Trending board. Mixing the strings is a 400: type youtube is invalid_request, source YouTube Trending is invalid_source. Omit limit and the board returns 25 rows. Omit percent_growth and youtube growth defaults to 12M.

Minecraft sat at search index 8 while four Minecraft titles occupied the board

The September 6, 2026 pull is the proof that the two YouTube surfaces do not rank the same object. get_growth with source youtube and keyword Minecraft closed at recent_value 8.0 on 2026-09-05 versus 9.0 on 2026-08-29, growth -11.11% on 7D, and 8.0 versus 8.0 on 2025-09-06, growth 0.0 on 12M, across 261 weekly points. The omitted-country YouTube Trending snapshot at as_of_ts 2026-09-06T06:01:36.749501+00:00 still carried four Minecraft titles in the default 25: rank 2 "1000 Players Simulate Mining Prison in Minecraft", rank 8 "Adopted by a SHARK FAMILY in Minecraft!", rank 11 "Simulating a Criminal Empire on Hardcore Minecraft", and rank 17 "I Found Villager Hospital In Minecraft". Search interest is a keyword index. The live board is a ranked list of video titles. A franchise can flood the board while its search index stays in single digits.

The clocks also differ. The youtube series last point is Saturday 2026-09-05. The board timestamp is the next morning. Do not treat a 7D search window and a live title list as the same hour.

LISA shows the matching case. Keyword LISA rose 143.75% on 7D, 16.0 on 2026-08-29 to 39.0 on 2026-09-05, and 95.0% on 12M from 20.0 on 2025-09-06. Rank 1 on the same omitted-country board was "LISA - SaWaDiKa (Official Music Video)". Keyword silo rose 80.95% on 7D, 42.0 to 76.0, and 375.0% on 12M from 16.0, with "SILO Season 4 Trailer Breakdown" at rank 7 and a second SILO title at rank 22. Alignment happens. It is not required.

Keyword history uses get_growth or get_time_series. The live board uses get_top_trends. REST mode names match the MCP tools. growth is an alias for get_growth. top_trends is an alias for get_top_trends. A REST top_trends call with type YouTube Trending and limit 3 on September 6, 2026 returned the same snapshot timestamp as the MCP omit-limit pull, with count 3.

{"mode": "get_growth", "source": "youtube", "keyword": "Minecraft", "percent_growth": ["7D", "12M"]}
{"mode": "get_time_series", "source": "youtube", "keyword": "mrbeast"}
{"mode": "get_top_trends", "type": "YouTube Trending"}
{"mode": "get_top_trends", "type": "YouTube Trending", "country": "GB", "limit": 5}

The MCP equivalents are get_growth(keyword="Minecraft", source="youtube", percent_growth=["7D", "12M"]), get_time_series(keyword="mrbeast", source="youtube"), and get_top_trends(type="YouTube Trending"). Docs for the YouTube search series and the YouTube Trending feed keep those contracts in one place each. This page is the split.

Omit percent_growth on mrbeast and get_growth returned a single 12M row: 45.0 on 2026-09-05 versus 36.0 on 2025-09-06, growth 25.0%, 261 points. Ask for both 7D and 1W on the same keyword and both rows print 45.0 versus 49.0 on 2026-08-29, growth -8.16%. Extra windows still sit on one source-plus-keyword request. See get_growth for the window list.

get_time_series for mrbeast on youtube returned weekly Saturday points from 2021-09-11 through 2026-09-05. Rows carried date, value, keyword, and source. No volume key. Index 100 printed twice in that span, on 2021-11-27 and 2024-07-20. The latest row was 45, matching the growth recent_value. Quote the series when a chart is required. Quote get_growth when the question is a percent.

get_top_trends with type youtube returned HTTP 400, error invalid_request, message Unknown type 'youtube', plus 24 valid feed labels ending in YouTube Trending. get_growth with source YouTube Trending and keyword mrbeast returned HTTP 400, error invalid_source, message 'youtube trending' is not a valid source., plus 16 valid sources: google search, google images, google news, google shopping, youtube, wikipedia, tiktok, reddit, amazon, news sentiment, news volume, app downloads, app rankings, steam, npm, python, all. The handler lowercased the rejected source. It did not map it to youtube.

The inverse fold exists on the feed name. type youtube trending on get_top_trends returned HTTP 200 and echoed type YouTube Trending, with the same omitted-country snapshot as the canonical label. llms.txt still describes feed labels as case-sensitive. The live handler accepted this lowercase YouTube string on September 6, 2026. Prefer the canonical YouTube Trending spelling in stored configs.

REST omitting type is a different 400. POST https://api.trendsmcp.ai/api with mode get_top_trends and limit 5, no type, returned HTTP 200 with envelope statusCode 400, error invalid_request, message The 'type' parameter is required, then the same 24-name list. llms.txt still names that miss missing_parameter. Read statusCode inside the JSON, not only the HTTP line. get_top_trends documents the default limit 25 and the credit math.

Title-string growth can print 2750 percent from a near-zero baseline

Pasting a board title fragment into source youtube is legal. It is often a thin series. Keyword SaWaDiKa on September 6, 2026 returned 7D growth 2750.0% from 2.0 on 2026-08-29 to 57.0 on 2026-09-05, 12M growth 999999.0 from baseline_value 0.0 on 2025-09-06, data_quality low, and the warning underlying signal is over 90% zero values; treat this trend with caution. 261 points still came back. Most of them were zeros. A 999999.0 print is the pipeline's answer to a zero baseline, not a demand figure to put in a forecast.

Artist and franchise keywords on the same pull stayed ordinary. Dragon Ball Super rose 35.71% on 7D, 14.0 to 19.0, with "Dragon Ball Super: Beerus - Official Super Surge Trailer (English Subtitles)" at rank 10. Ariana Grande rose 2.56% on 7D, 39.0 to 40.0, with "FOCKER IN-LAW Trailer 2 (2026) Ariana Grande, Ben Stiller" at rank 4. Those are search phrases people already type. Title tokens that barely existed last year are the ones that trip the zero-baseline percent.

Omit limit returns 25. Country and rank_change change the board.

get_top_trends with type YouTube Trending and no limit returned limit 25, offset 0, count 25. Rank 1 was the LISA video. That omitted-country snapshot is what MCP returned for the canonical type, the lowercase type, the REST top_trends alias, and sort rank_change. The timestamp stayed 2026-09-06T06:01:36.749501+00:00.

country GB with limit 5 did not reuse that snapshot. as_of_ts moved to 2026-09-06T06:07:44.645634+00:00, the payload added "country": "GB", and rank 1 became "Tzanca Uraganu - Cum e sa-ti pacalesti inima [videoclip oficial] 2026". Rank 2 was a Strictly Come Dancing trailer. The Minecraft prison video fell to rank 3. LISA sat at rank 4. YouTube Trending is one of the feeds that accepts an ISO country code. Omit country for the default stored board. Pass GB when the question is Britain. The two lists are not a sort of the same 25 titles.

sort rank_change and window 7d with limit 10 returned count 2, baseline_ts 2026-08-30T02:01:37.593716+00:00, and two mover objects. Rank 101 "GTA V LIVE #shortsfeed #shortlive #gta5 #shorts #live" climbed 91 positions from prev_rank 192. Rank 15 "Jarahn x SHLD - Yu Les (Audio) feat. Nene Morus & Rich Antony" climbed 12 from prev_rank 27. limit is a maximum. A quiet mover window can return fewer rows than requested. Ranks 101 and 192 also show the stored board is deeper than the default 25-row payload. Page with offset on a paid plan when the job is the long tail. The YouTube Trending versus Google Trends page is the cross-platform follow-up. This split stays inside YouTube: search keyword versus live title.