npm weekly downloads, mobile app install growth, and web traffic trends are the clearest signals of developer and user adoption. Trends MCP delivers all of them to your AI in one query - useful for technology due diligence, framework selection, and developer market research.
Updated 2026-08-19
Developer ecosystem trend research is a diligence loop: pick a package, app, or tool, pull weekly adoption history of about five years on a 0-100 index, then compare 1 month, 3 month, and 6 month growth against the next-best alternative. As of August 2026, one MCP or REST API covers 25+ platforms, so npm downloads, Play install trends, Steam players, branded search, and site traffic share one auth and one response shape. The free plan is 100 requests per month. Starter is $19 per month. No app-store developer account is required for the public install-trend path.
Adoption without a vendor briefing is read from public install and download series first, then from search and traffic, then from whatever the company claims in a deck, because npm weekly downloads are the library-level tell and a compounding year of weekly points often shows mindshare before revenue is obvious. A package that peaked two years ago can still look busy in a single month of CI traffic.
The first call on a named library is weekly history, because percentage growth on a tiny base is a trap. The 0-100 index shows where the series sits relative to itself. Growth windows of 1 month, 3 months, and 6 months then show whether the latest print is a release week or a slope. Ranked npm lists help when the question is "what is rising in this category" rather than "is this one package alive."
Search and docs traffic catch the buyer who never installs. A CLI can be widely downloaded in CI while the commercial console is stalling. Pairing npm with web traffic data for the product domain stops that miss. Google Search on the product name is the awareness check for people who have not picked a package yet.
Package comparison works when each name is queried with the same windows and the same 0-100 index, then laid out as a table of growth rates rather than as a social-media argument, which is how pairs such as React versus Vue versus Svelte or OpenAI SDK versus Anthropic SDK get ranked in one sitting. The assistant can run each get_growth call and return ranks. Humans still decide what "winning" means for the repo.
Downloads are not quality. A deprecated compatibility shim can grow because it is a transitive dependency. A beloved library can shrink because the ecosystem moved to a successor under a new name. Always check whether the keyword is the package people actually install. Scoped names such as @anthropic-ai/sdk matter. A generic token such as sdk does not.
Five years of weekly points is long enough to see a framework generation. It is also long enough to see hype cycles that already resolved. Diligence that only reads 30 days will call a launch spike a takeover. Diligence that only reads five years will miss a new default that appeared this quarter. Use both.
App, Play, and Steam signals sit beside npm when the product is a client, a game, or a consumer wrapper around a developer platform, because those sources answer user adoption rather than GitHub stars and they share the same 0-100 index as package downloads. App download trends use Google Play install-trend estimates (via AppBrain) queried by Android bundle ID, for example com.openai.chatgpt. Steam trends use concurrent player series for titles where the store page is the market.
A developer tool with a mobile companion can be winning npm while the app is flat, or the reverse. Gaming platforms and overlay tools show up on Steam before they show up as a JavaScript package. Live top lists for Google Play and the App Store are the greenfield scan when the question is which mobile developer utilities are charting right now.
No store credentials are part of this path. That is the point for analysts who cannot join a Play Console. Estimates are directional. They are not audited install invoices. Treat them as a slope, not as a finance line.
Diligence teams time lookups to decision gates such as screening, partner review, and a final memo, not to an always-on firehose, so screening uses ranked growth, partner review adds weekly history plus traffic or search, and the memo adds competing names on the same windows.
Quota follows that cadence. As of August 2026, 100 requests per month on the free plan covers a light screening roster. Starter at $19 per month covers a busier pipeline of package and app checks. Repeat the same identifiers. Do not invent new spellings each week or the history will not line up.
Technology research for investment uses the same calls with a different last sentence. Investors ask whether a platform is actually being adopted. Engineering managers ask whether the team should standardize on it. Both need slope plus level. Only one needs a cap table.
Public ecosystem data misses private registries, most GitHub clone counts, and any install that never hits npm or Play, so the honest output for an unobserved stack is "not observed" rather than a guessed share of developer mindshare on a 0-100 index. PyPI and other language ecosystems may be available as separate sources on the same 25+ platform API.
AppBrain-style Play estimates can lag and can be noisy below a meaningful install base. Steam concurrent players move with sales, streams, and seasonals. Web traffic estimates are directional for larger sites. None of these series replace security review, license review, or a reading of the actual changelog.
Stars, Hacker News points, and a single viral demo are not substitutes for weekly downloads. They are useful as event markers when overlaid on the five-year series. The tools below are the calls that diligence loop uses.
Tools for this workflow
get_trendsChart npm weekly downloads for any package over 5 years - the clearest signal of whether a library or framework is gaining or losing developer adoption.
get_trends(keyword='@anthropic-ai/sdk', source='npm', data_mode='weekly')get_growthCompare competing packages or frameworks side by side in one call - npm downloads, web traffic, and Google Search growth together reveal which technology is winning developer mindshare.
get_growth(keyword='@anthropic-ai/sdk', source='npm', percent_growth=['1M', '3M', '6M'])get_ranked_trendsFind the fastest-growing npm packages this year - rank by YoY download growth to spot emerging libraries before they become the default choice in their category.
get_ranked_trends(source='npm', sort='yoy_pct_change', limit=30)get_top_trendsSee which apps and developer tools are topping Google Play and App Store charts right now - surfaces emerging mobile developer tools and platforms gaining adoption.
get_top_trends(type='Google Play', limit=20)FAQ