Today's story: Claude Code
Coding agents outrun attention as installs plateau
Claude Code's rise shows developer tools winning on talk, not downloads, while AI APIs surge.
The numbers: Claude Code is today's biggest mover: up 2 places to No. 3, Hacker News stories at 1.52× the four-week pace. Claude Code: 1.23× momentum across 2 signals. 5 of 21 trends are climbing and 6 are cooling. SpaceX Starship still leads the board at 1.38× (day 8).
Claude Code, signal by signal
Each pedestal is one signal: its height is last week's typical day against the four weeks before, and the ring marks that normal. A lit pedestal is rising 1.15× or more.
Claude Code is today's biggest mover: up 2 places to No. 3, Hacker News stories at 1.52× the four-week pace.
Claude Code: 9.6 Hacker News stories a day on a typical day last week, against 6.3 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 10/07, HN 10/07. Source: Hacker News search (Algolia), story titles.
Claude Code climbed two places to No. 3 today, driven by Hacker News chatter at 1.52× its four-week pace, even as daily installs slipped to 0.99× (2.2M vs 2.3M). The gap between attention and adoption—installs at 0.99×, stories at 1.52×—marks a split: developers are discussing the tool more than they are installing it. Meanwhile, AI API SDKs hit 21M daily installs (1.19×), and OpenAI Codex grew 116× year-over-year to 3.4M/day, showing raw usage is shifting to lower-level building blocks.
The trend is not a single tool but a maturing coding-agent stack. Claude Code's momentum comes from its role in a workflow where AI APIs and SDKs are the plumbing—installs there dwarf any single agent. OpenAI Codex's Wikipedia interest jumped 2.8×, a sign of broader curiosity beyond the terminal. As agents become the interface, the underlying APIs become the commodity. The plateau in Claude Code installs suggests the agent market is saturating at the top, while the API layer keeps compounding.
Contrast this with cooling hardware bets: humanoid robots, smart glasses, and robotaxis all fell in Wikipedia attention (0.82×, 0.72×, 0.71×). The board's center of gravity has shifted from physical machines to software that writes software. Starship still leads for an eighth day, but its Hacker News score is a low 0.23×, meaning it's a spectacle, not a developer movement. The real story is that coding agents are becoming the new platform, and the numbers show it's a conversation-led climb, not an install-led one.
Watch next: Watch whether Claude Code's installs catch up to its Hacker News momentum within a week, or if API SDK growth pulls developers away from standalone agents.
Column drafted by DeepSeek-V4-Flash from the numbers below; every figure checked against them.
The numbers
Hacker News carried 9.6 stories a day with "Claude Code" in the title last week, against 6.3 a day in the four weeks before (1.52×). On a typical day last week, npm installs of @anthropic-ai/claude-code reached 2.2M, 0.99× the 2.3M of a typical day in the four weeks before.
Still No. 1 for day 8: SpaceX Starship at 1.38×, with Wikipedia readers at 1.38× (1,901 on a typical day against 1,383).
Talk ahead of installs: attention to Claude Code runs at 1.52× while installs sit at 0.99×. Watch whether usage follows.
Cooling fastest: Robotaxis at 0.71×, with Wikipedia readers down to 175 on a typical day from 245. Part of that is a spike on Sep 8 in the weeks before.
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 StarshipPasskeysClaude CodeRobotaxis
Talk ahead of installsBroad riseCooling day5 of 21 trends are climbing; SpaceX Starship leads at 1.38×
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 10/07, HN 10/07. 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 10/07, HN 10/07. Source: npm downloads API, Wikimedia pageviews, Hacker News search.
SpaceX Starship draws the most Wikipedia readers against its own normal: 1.38× over the last week
Each cell is one day's readers divided by that trend's four-week typical day. Brighter runs hot, darker runs cold.
Data through npm 10/05, Wikipedia 10/07, HN 10/07. Source: Wikimedia pageviews API, user agents only.
Every number on the board
| SpaceX Starship | Space | 1.38× | Climbing | – | – | 1,901 | 1.38× | 0.1 | 0.23×* | 0 |
|---|---|---|---|---|---|---|---|---|---|---|
| Passkeys | Web platform | 1.3× | Climbing | 1.1M | 1.3× | 93 | 1.05×* | 0.1 | 0.4×* | 0 |
| Claude Code | AI coding | 1.23× | Climbing | 2.2M | 0.99× | – | – | 9.6 | 1.52× | 0 |
| AI API SDKs | AI agents | 1.19× | Climbing | 21M | 1.19× | – | – | – | – | 0 |
| OpenAI Codex | AI coding | 1.18× | Climbing | 3.9M | 1.11× | 110 | 2.8×* | 5.3 | 1.24× | 0 |
| Model Context Protocol | AI agents | 1.14× | Steady | 12M | 1.27× | 1,663 | 1.09× | 7.3 | 1.06× | 0 |
| WebGPU | Web platform | 1.14× | Steady | 1.7M | 1.17× | 142 | 1.12× | 1.3 | 1.56×* | 0 |
| Starlink | Space | 1.1× | Steady | – | – | 2,195 | 1.1× | 0.7 | 1.54×* | 0 |
| Small modular reactors | Energy | 1.09× | Steady | – | – | 510 | 1.09× | 0.1 | n/a* | 0 |
| AI agents | AI agents | 1.05× | Steady | 782k | 1.26× | 962 | 0.81× | 13 | 1.15× | 4 |
| Heat pumps | Energy | 1.05× | Steady | – | – | 502 | 1.05× | 0.0 | 0×* | 1 |
| Bun runtime | Web platform | 1.03× | Steady | 708k | 1.08× | 240 | 0.98× | 0.7 | 0.83×* | 0 |
| Solid-state batteries | Energy | 1.03× | Steady | – | – | 424 | 1.03× | 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 | 0.89× | 0.4 | 0.63×* | 0 |
| eVTOL air taxis | Space | 0.86× | Cooling | – | – | 169 | 0.86× | 0.0 | n/a* | 0 |
| Vibe coding | AI coding | 0.83× | Cooling | – | – | 1,755 | 1.13× | 0.7 | 0.61× | 1 |
| Quantum computing | Devices | 0.83× | Cooling | – | – | 1,700 | 0.93× | 2.3 | 0.74× | 0 |
| Humanoid robots | Robots | 0.82× | Cooling | – | – | 301 | 0.82× | 0.9 | 1.14×* | 0 |
| Smart glasses | Devices | 0.72× | Cooling | – | – | 153 | 0.72× | 0.7 | 1.05×* | 1 |
| Robotaxis | Robots | 0.71× | Cooling | – | – | 175 | 0.71× | 0.9 | 2.2×* | 1 |
| Vehicle-to-grid | Energy | n/a | Not enough data | – | – | 61 | 0.98×* | 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/2026-10-08.json.