Protocol adoption outpaces model growth
MCP and AI API SDKs climb as install velocity decouples from model hype, signaling a move toward stable integration layers.
The numbers: SpaceX Starship is today's fastest-rising trend: Wikipedia readers at 1.57× the four-week pace. Momentum across 1 signal: 1.57×. 4 of 21 trends are climbing and 4 are cooling. SpaceX Starship has led 7 days running.
SpaceX Starship: 2,165 Wikipedia readers a day on a typical day last week, against 1,378 the four weeks before
Bars are single days, the line is the 7-day average, the shaded band is the last 7 days and the dashed line is the four-week typical day (median).
Data through npm 10/05, Wikipedia 10/06, HN 10/06. Source: Wikimedia pageviews API, user agents only.
Model Context Protocol installs reached 12M/day versus 9.7M previously, a 1.27× jump. AI API SDKs followed at 21M/day, up 1.19× from 17M. These gains occur while Claude Code installs dipped to 2.2M/day from 2.3M. The divergence suggests developers are prioritizing connection standards over specific coding assistants. This pattern indicates a maturation phase where interoperability matters more than individual tool features.
Attention metrics confirm this trend. MCP saw 7.7 Hacker News stories daily, up 1.14× from 6.8. Wikipedia views for MCP rose to 1,663/day, a 1.11× increase. Meanwhile, Claude Code attention jumped 1.3× to 8.4 stories, but its install growth stalled at 0.99×. The gap between buzz and actual usage highlights a preference for durable protocols that persist across different model releases and vendor changes.
This trajectory contrasts with cooling sectors like Quantum computing, which fell to 0.71× momentum. The rise of MCP and SDKs points to a builder focus on reliable plumbing. As OpenAI Codex grows to 3.4M/day, the surrounding ecosystem of protocols becomes the critical differentiator. Developers are betting on the ability to swap models without rewriting integration logic, making the protocol layer the new center of gravity.
Watch next: Monitor whether MCP install velocity sustains above 1.1× as new model releases occur.
Column drafted by qwen/qwen3.8-27b from the numbers below; every figure checked against them.
The numbers
The Wikipedia article "SpaceX Starship" drew 2,165 readers on a typical day last week, 1.57× its four-week norm of 1,378. The four weeks before included a spike on Sep 28.
Next on the board: Passkeys at 1.3×, led by installs at 1.3× (1.1M on a typical day against 807k).
Talk ahead of installs: attention to Claude Code runs at 1.3× while installs sit at 0.99×. Watch whether usage follows.
Cooling fastest: Smart glasses at 0.68×, with Wikipedia readers down to 153 on a typical day from 224.
The long view: OpenAI Codex installs over the last four weeks are 116× the same four weeks a year earlier (3.4M a day against 29k).
Trend pages: SpaceX StarshipPasskeysAI API SDKsSmart glasses
Talk ahead of installs4 of 21 trends are climbing; SpaceX Starship leads at 1.57×
Momentum is a typical day (median) of the last 7 divided by a typical day of the 28 before, averaged across every signal with enough data. 1× means no change. Rings mark each signal.
Data through npm 10/05, Wikipedia 10/06, HN 10/06. Source: npm downloads API, Wikimedia pageviews, Hacker News search.
5 of 7 developer tools are being installed faster than they are being read about
Across: attention momentum (Wikipedia readers and Hacker News stories). Up: install momentum (npm). Both on log scales; the 1× lines split the corners.
Data through npm 10/05, Wikipedia 10/06, HN 10/06. Source: npm downloads API, Wikimedia pageviews, Hacker News search.
SpaceX Starship draws the most Wikipedia readers against its own normal: 1.57× over the last week
Each cell is one day's readers divided by that trend's four-week typical day. Orange runs hot, dark runs cold.
Data through npm 10/05, Wikipedia 10/06, HN 10/06. Source: Wikimedia pageviews API, user agents only.
Every number on the board
| SpaceX Starship | Space | 1.57× | Breakout | – | – | 2,165 | 1.57× | 0.1 | 0.23×* | 0 |
|---|---|---|---|---|---|---|---|---|---|---|
| Passkeys | Web platform | 1.3× | Climbing | 1.1M | 1.3× | 92 | 1.04×* | 0.1 | 0.36×* | 0 |
| AI API SDKs | AI agents | 1.19× | Climbing | 21M | 1.19× | – | – | – | – | 0 |
| Model Context Protocol | AI agents | 1.17× | Climbing | 12M | 1.27× | 1,663 | 1.11× | 7.7 | 1.14× | 0 |
| Claude Code | AI coding | 1.14× | Steady | 2.2M | 0.99× | – | – | 8.4 | 1.3× | 0 |
| OpenAI Codex | AI coding | 1.13× | Steady | 3.9M | 1.11× | 99 | 2.5×* | 5.0 | 1.15× | 0 |
| Small modular reactors | Energy | 1.12× | Steady | – | – | 527 | 1.12× | 0.0 | n/a* | 0 |
| WebGPU | Web platform | 1.11× | Steady | 1.7M | 1.17× | 137 | 1.06× | 1.1 | 1.6×* | 0 |
| Starlink | Space | 1.1× | Steady | – | – | 2,195 | 1.1× | 0.7 | 1.43×* | 0 |
| AI agents | AI agents | 1.08× | Steady | 782k | 1.26× | 962 | 0.81× | 14 | 1.24× | 0 |
| Bun runtime | Web platform | 1.06× | Steady | 708k | 1.08× | 244 | 1.03× | 1.0 | 1.22×* | 0 |
| Solid-state batteries | Energy | 1.06× | Steady | – | – | 427 | 1.06× | 0.0 | 0×* | 0 |
| Heat pumps | Energy | 1.05× | Steady | – | – | 502 | 1.05× | 0.0 | 0×* | 0 |
| Gemini CLI | AI coding | 1.02× | Steady | 66k | 1.02× | – | – | 0.0 | n/a* | 0 |
| Local LLMs | AI agents | 1× | Steady | 192k | 1.13× | 634 | 0.9× | 0.7 | 1.11×* | 0 |
| Vibe coding | AI coding | 0.95× | Steady | – | – | 1,755 | 1.16× | 0.9 | 0.77× | 0 |
| eVTOL air taxis | Space | 0.91× | Steady | – | – | 181 | 0.91× | 0.0 | 0×* | 0 |
| Humanoid robots | Robots | 0.81× | Cooling | – | – | 301 | 0.81× | 0.9 | 1.04×* | 0 |
| Quantum computing | Devices | 0.71× | Cooling | – | – | 1,700 | 0.93× | 1.7 | 0.55× | 0 |
| Robotaxis | Robots | 0.7× | Cooling | – | – | 175 | 0.7× | 0.9 | 2.2×* | 0 |
| Smart glasses | Devices | 0.68× | Cooling | – | – | 153 | 0.68× | 0.7 | 1×* | 0 |
| Vehicle-to-grid | Energy | n/a | Not enough data | – | – | 54 | 0.87×* | 0.0 | 0×* | 0 |
Method and sources
Every trend gets up to three free, daily signals. Installs: npm registry downloads for the packages named on each trend page (developer adoption; it counts CI and mirrors too, so it measures direction, not people). Attention: English Wikipedia readers per day, human agents only, and Hacker News stories with the term in the title. For each signal we divide a typical day (the median) of the last 7 complete days by a typical day of the 28 days before. Medians keep one launch-day spike from carrying a whole week; both windows are whole weeks, so weekend dips sit equally in both. Momentum is the geometric mean of the signals that clear a minimum base. npm reports some days as zero for every package at once; we treat those as missing.
The column is drafted by a free language model from the day's computed facts only, and is published only if every number in it matches those facts; otherwise the computed paragraphs stand alone. Theme labels come from Jev (or its Cloudflare Clef fallback, as noted) choosing from a fixed list of ten.
Stages: Breakout 1.5× and up, Climbing 1.15×, Steady 0.87× to 1.15×, Cooling below. Raw data: /api/trends/2026-10-07.json.