Pass source news sentiment for tone, or source news volume for article counts. There is no sentiment or volume live board.
Live data as of 2026-09-23
A news sentiment vs news volume comparison on Trends MCP is two keyword sources on get_growth and get_time_series. Source news sentiment returns a 0-100 tone series for any phrase. Source news volume returns a 0-100 mention-volume series plus an article-count volume field when the pipeline has it. There is no get_top_trends type for either string. On September 23, 2026, tesla closed at 54.7 on news sentiment and 0.2 on news volume with 49 articles, recent_date 2026-09-19. google news for the same name sat at 12.0. A comma-separated get_growth call on both sources returned statusCode 502. type news sentiment is a 400.
The September 23, 2026 Google News Top News board is the split that still matters next month: live rank is a headline list, news sentiment can rise while news volume prints a near-floor close, and google news can sit in the middle with no article count. Omitting limit returned limit 25, count 25, offset 0, as_of_ts 2026-09-23T06:01:40.600896+00:00. Rank 19 named Qualcomm. Rank 20 named Apple. Rank 25 named Taylor Swift.
| Token | Kind | news sentiment 7D | news volume 7D | google news 7D |
|---|---|---|---|---|
| tesla | off-board | 54.7 from 47.2, +15.89% | 0.2 from 9.4, 49 from 1,913 articles, -97.87% | 12.0 from 16.0, -25.0%, no volume field |
| nvidia | off-board | 72.0 from 67.6, +6.51% | CloudFront HTTP 504 | 7.0 from 8.0, -12.5% |
| meta | off-board | 72.9 from 64.5, +13.02% | CloudFront HTTP 504 | 10.0 from 13.0, -23.08% |
| Apple | rank 20 name | 70.8 from 56.2, +25.98%, close 2026-09-12 | CloudFront HTTP 504 | 43.0 from 63.0, -31.75%, close 2026-09-12 |
| Taylor Swift | rank 25 name | CloudFront HTTP 504 | not pulled | not pulled |
Live rank is not a sort of either index. tesla was off the board and still printed the only complete three-source close on this pull: sentiment 54.7, volume 0.2 with 49 articles, google news 12.0, same recent_date 2026-09-19, 261 points on each dedicated call that returned.
The volume close is the health check, not the coverage story. tesla news volume 7D was 0.2 from 9.4, -97.87% index and 49 articles from 1,913, -97.44% volume. 30D was 0.2 from 9.7, 49 from 1,986. 12M was 0.2 from 24.8, 49 from 5,075. YTD was 0.2 from 21.3, 49 from 4,358. news sentiment on that same recent_date rose: 54.7 from 47.2 on 7D, then 54.7 from 76.9 on 30D, -28.87%, then 54.7 from 34.5 on 12M, +58.55%, then 54.7 from 74.7 on YTD, -26.77%. google news sat at 12.0 from 16.0 on 7D, 12.0 from 100.0 on 30D, -88.0%, 12.0 from 9.0 on 12M, +33.33%. Quote 54.7 next to 0.2 next to 12.0. A 49-article Tesla close beside a still-live google news series is a source-health flag.
nvidia is the cleaner tone-versus-news-search join because volume missed. Sentiment closed 72.0 from 67.6 on 7D, 72.0 from 60.9 on 30D, +18.23%, 72.0 from 54.6 on 12M, +31.87%, 72.0 from 54.8 on YTD, +31.39%. google news closed 7.0 from 8.0 on 7D, 7.0 from 80.0 on 30D, -91.25%, 7.0 from 2.0 on 12M, +250.0%. Tone rose over 30D while the news-search index fell from 80.0 to 7.0. Dedicated news volume for nvidia returned CloudFront HTTP 504 HTML. Quote the miss.
Apple lagged a week. Sentiment closed 70.8 from 56.2 on 7D with recent_date 2026-09-12, 262 points. google news closed 43.0 from 63.0 on the same 2026-09-12 stamp, 261 points. tesla on this morning's pull had already moved to 2026-09-19. Store the close date per keyword. Do not assume one recent_date for the whole news cluster.
REST omitting percent_growth on google news tesla defaulted to 12M and printed 12.0 from 9.0, +33.33%, 261 points. 1W matched 7D exactly on that source: 12.0 from 16.0, -25.0%. 14D matched 2W: 12.0 from 75.0 on 2026-09-05, -84.0%. 1M matched 30D: 12.0 from 100.0, -88.0%. 1Y matched 12M. The matching omit on news sentiment and news volume returned HTTP 504 on this run. Quote those timeouts. Do not invent a 12M sentiment default from a 504.
Two strings, one tool family. REST uses mode get_growth or get_time_series. MCP uses the matching tool names. Tone windows live on the news sentiment data page. Article-count windows live on news volume data. Window defaults live on get-growth. The volume-versus-headline-board split is documented with news volume vs Google News Top News.
{"mode": "get_growth", "source": "news sentiment", "keyword": "tesla", "percent_growth": ["7D", "30D", "12M", "YTD"]}
{"mode": "get_growth", "source": "news volume", "keyword": "tesla", "percent_growth": ["7D", "30D", "12M", "YTD"]}
{"mode": "get_growth", "source": "news sentiment, news volume", "keyword": "tesla", "percent_growth": ["7D", "30D", "12M"]}
{"mode": "get_time_series", "source": "news sentiment", "keyword": "tesla"}
{"mode": "get_top_trends", "type": "Google News Top News"}
The MCP equivalents are get_growth(keyword="tesla", source="news sentiment", percent_growth=["7D", "30D", "12M", "YTD"]), get_growth(keyword="tesla", source="news volume", percent_growth=["7D", "30D", "12M", "YTD"]), and get_top_trends(type="Google News Top News"). REST top_trends with type Google News Top News and limit 3 on September 23, 2026 returned the same snapshot timestamp as the omitted-limit MCP pull, with count 3: the BBC Iran row, the Politico press-ban row, and the CBS Greenland row.
The comma-separated tesla call is the timeout check. MCP get_growth with source news sentiment, news volume returned statusCode 502, error all_sources_unavailable, sources_requested ["news sentiment", "news volume"], sources_successful 0. Each source_results row printed status unavailable and message Source did not respond in time. Dedicated single-source tesla calls then returned 200. Store the join as two requests when the comma path 502s.
get_time_series on news sentiment, news volume returned HTTP 200 with envelope statusCode 400, error invalid_source, message 'news sentiment, news volume' is not a valid source., plus 16 names and no all. Multi-source strings belong on get_growth only. Source News Sentiment case-folded and returned HTTP 200, echoing news sentiment, tesla 7D 54.7 from 47.2, +15.89%, 261 points. Source NEWS VOLUME returned HTTP 504 on this run. Store the lowercase source strings.
meta sentiment closed 72.9 from 64.5 on 7D, +13.02%, then 72.9 from 34.0 on 30D, +114.41%, then 72.9 from 78.8 on 12M, -7.49%. google news for meta closed 10.0 from 13.0 on 7D, 10.0 from 100.0 on 30D, -90.0%, 10.0 from 5.0 on 12M, +100.0%. Dedicated news volume for meta returned CloudFront HTTP 504. Taylor Swift on news sentiment also 504ed. Quote the misses. Do not invent an article count for those names.
get_top_trends with type news sentiment returned HTTP 200 with envelope statusCode 400, error invalid_request, message Unknown type 'news sentiment', plus 24 valid feed labels that include Google News Top News and do not include any sentiment board. type News Sentiment failed the same way. type news volume failed the same way. The handler did not map any of those three strings to the keyword sources.
type google news top news on get_top_trends returned HTTP 200 and echoed type Google News Top News, with the same as_of_ts 2026-09-23T06:01:40.600896+00:00 and the same first three headlines as the canonical label. type google news returned invalid_request. llms.txt still describes feed labels as case-sensitive. The live handler accepted the lowercase Top News alias on September 23, 2026. Prefer the canonical Google News Top News spelling in stored configs. Do not store news sentiment as a feed type, and do not expect google news to get the same mapping.
get_growth with source Google News Top News and keyword tesla returned HTTP 200 with envelope statusCode 400, error invalid_source, message 'google news top news' 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 news volume or google news. The get_time_series reject list for a comma-separated pair is the same 16 names without all.
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_time_series on tesla news sentiment returned HTTP 504 on this run. Peak dates for that series are therefore not on this page. google news tesla still printed 261 weekly growth points with recent_date 2026-09-19. Follow-up for a headline board stays on type Google News Top News. Follow-up for a search-interest index stays on source google news at the Google News keyword trends API.
get_top_trends MCP pricing is 1 credit per 10 requested rows, max 10 credits per call. REST counts 1 request per call. Free boards still cap at the top 10. The 25-row Google News Top News pull on this page is the paid default, which is also what the handler returns when limit is omitted on a paid key. Tone follow-up is the news sentiment data page. This split stays inside news: coverage tone versus article count for one phrase.
FAQ