This week in trends · Jul 18–24, 2026 · 6 min read

What's trending this week — the measured top, Jul 18–24

The measured top of the week from our live radar: what actually gained speed between Jul 18 and Jul 24, on which platforms, with every number carrying its unit and measurement date.

Published Jul 24, 2026 · by the Lolik team, from our own radar data

Key takeaways

  • Agent tooling is the week's real story — three of the five confirmed breakouts (Agent-Reach, firecrawl, browser-use) are infrastructure for AI agents.
  • Durability beat raw size: the top spot goes to 402 stars/day sustained for 20 days, not to the 51K views/day spike that is only 2 days old.
  • 338 new signals entered tracking this week; only 44 of 898 earned the breakout label — most of what looks hot never confirms.
  • All figures measured Jul 24, 2026 on our own radar — units and tracking windows stated inline.

898

signals live-tracked this week

338

new signals first seen this week

44

in breakout phase

7

niches covered

Where the 898 live signals stand

Decaying 662Watch 192Breakout 44

Most of what spikes never confirms — only 44 of 898 signals held the breakout label this week. Counts measured Jul 24, 2026.

The confirmed top 5 — signals with sustained, observed velocity

These five cleared our confirmation bar: velocity sustained across enough tracked days to be marked observed, not just spiking. Ordered by overall score (velocity × acceleration × durability), not by raw numbers alone. The bar under each velocity is scaled within its own unit group — stars/day and views/day are never compared on one scale.

#1

Agent-Reach

BreakoutBusiness / AI tooling
402 stars/dayGitHub · tracked 20 days · first seen Jul 4
A CLI that gives AI agents read access to Twitter, Reddit, YouTube, GitHub and more without per-API fees. Twenty days of sustained star velocity is the signature of a real developer wave, not a launch-day spike.
Full tracked history
#2

HeyGen (AI video)

BreakoutAI
15,978 views/dayYouTube · tracked 10 days · first seen Jul 14
AI avatar video content keeps compounding on YouTube — this signal accelerated hard mid-week (+450 views/day of added velocity at measurement). The topic has carried our AI niche for two weeks straight.
Watch on YouTube: HeyGen (AI video)
#3

Vibe-Trading

BreakoutFinance
238 stars/dayGitHub · tracked 24 days · first seen Jun 25
"Your personal trading agent" — the vibe-coding wave crossing into retail finance. Still climbing after three weeks on the radar, which is unusually durable for a finance repo.
Full tracked history
#4

firecrawl

BreakoutAI
187 stars/dayGitHub · tracked 28 days · first seen Jun 25
The web-scraping API for AI agents refuses to cool off — four straight weeks above 170 stars/day. The agent-infrastructure layer keeps being the steadiest demand pocket in AI.
Full tracked history
#5

browser-use

BreakoutAI
169 stars/dayGitHub · tracked 28 days · first seen Jun 25
Same story, adjacent layer: making websites operable by agents. Together with firecrawl and Agent-Reach, three of this week's five confirmed breakouts are agent-tooling — that cluster is the week's real headline.
Full tracked history

The early watchlist — flagged, still confirming

Everything below is under a week old on the radar. This is the phase where the timing edge actually lives — and also the phase with the most false starts, which is why we label it honestly instead of shouting "breakout" on day two. How to read this phase: our guide to spotting trends early.

#1

"The WORST Part About Being Rich"

WatchFinance
28,874 views/dayYouTube · tracked 4 days · first seen Jul 21
Money-psychology content with strong mid-week acceleration. Four days of data is enough to flag, not enough to confirm — that is exactly what the watch phase means.
Watch on YouTube: "The WORST Part About Being Rich"
#2

"Why Your Coaching Business Has a Churn Problem"

WatchBusiness
17,102 views/dayYouTube · tracked 7 days · first seen Jul 18
Creator-economy business content holding a full week of elevated velocity — the retention/churn angle is resonating with the coaching audience right now.
Watch on YouTube: "Why Your Coaching Business Has a Churn Problem"
#3

"Gear + Dirty Bulking Is Insane"

WatchHealth
51,782 views/dayYouTube · tracked 2 days · first seen Jul 23
The single fastest raw velocity on the radar this week — but only two days old, so it is still confirming. Big number, small sample: we label that honestly rather than calling it a breakout.
Watch on YouTube: "Gear + Dirty Bulking Is Insane"

What we'd do with this, practically

If you make content for the AI/developer audience, the agent-tooling cluster is a confirmed, durable wave — comparison and tutorial angles still have room while the topics climb. The finance-psychology and coaching-churn signals suit creators in those niches who can publish within days, not weeks: watch-phase topics reward speed and punish delay.

For the full pipeline from a signal like these to an actual brief — finding topics before they peak and monetizing them realistically.

Measured, in public

In our own backtest, early-phase alerts kept growing over the next 5 days in 38.8% of cases vs a 5.3% baseline across every tracked signal-day — 7.3× more likely to keep growing than a random tracked trend (n=165, Jul 2–23, 2026). An independent window with zero overlap (Jul 13–23, 2026) reproduced the lift at 8.6× (n=75).

Point-in-time figures, re-measured weekly — full methodology and every call, wins and misses, in the public log.

Common questions

How is “trending” measured here?

Every signal is tracked daily on its native unit — stars/day for GitHub repos, views/day for YouTube — and ranked by a score that weighs velocity, acceleration and durability. All figures in this post were measured on Jul 24, 2026; they are point-in-time snapshots, not live numbers.

What do Breakout and Watch mean?

Watch = elevated velocity, still confirming (typically under a week of data). Breakout = velocity sustained and accelerating across enough tracked days to be confirmed. Fresh spikes stay in Watch no matter how big the raw number is — a 2-day spike can die as fast as it appeared.

Why didn't the biggest number win the top spot?

Because raw size and confidence are different things. A 51K views/day spike with 2 days of history is less reliable than 402 stars/day sustained for 20 days. The ranked list only includes signals our system marks as observed; everything younger sits in the watchlist.

Can I see these signals live instead of weekly?

Yes — the radar updates daily and the free tools run on the same data, no signup. The weekly post is the measured summary; the product is the live feed. Timing edge is a possibility, not a guarantee — outcomes depend on your execution.

Want next week's list while it's still forming?

The free tools run on the same live data as the paid radar — no signup, no card. Start there; upgrade only if the timing edge earns it.