<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: MAT Armel</title>
    <description>The latest articles on DEV Community by MAT Armel (@alone225).</description>
    <link>https://dev.to/alone225</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F1753538%2F48f3f816-a289-4d6e-b507-5b2c186e2b21.jpg</url>
      <title>DEV Community: MAT Armel</title>
      <link>https://dev.to/alone225</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/alone225"/>
    <language>en</language>
    <item>
      <title>Fake GitHub repos are buying stars in batches to climb trending lists</title>
      <dc:creator>MAT Armel</dc:creator>
      <pubDate>Sat, 21 Nov 2026 17:57:21 +0000</pubDate>
      <link>https://dev.to/alone225/fake-github-repos-are-buying-stars-in-batches-to-climb-trending-lists-1cp3</link>
      <guid>https://dev.to/alone225/fake-github-repos-are-buying-stars-in-batches-to-climb-trending-lists-1cp3</guid>
      <description>&lt;p&gt;Earlier this week, a repo called "google-ultra" was #1 in Python on my own site. It promised paid Google AI tiers for free. I blocked it in the morning, and GitHub took it down within a day.&lt;/p&gt;

&lt;p&gt;It had climbed my ranking the way the rest of these campaigns do: with stars delivered in batches, to brand-new accounts, on the same days. This post is what I've seen this month, with the numbers, the mistakes my filters made, and how you can spot these repos yourself.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why a growth ranking is a target
&lt;/h2&gt;

&lt;p&gt;I build &lt;a href="https://upstar.dev" rel="noopener noreferrer"&gt;Upstar&lt;/a&gt; alone, from Abidjan. It ranks the GitHub repos that are taking off in 69 topics, every hour. The score is the stars a repo gained in the period, divided by the square root of the stars it had before. A small project that doubles in a week can beat a giant that gains 2,000.&lt;/p&gt;

&lt;p&gt;That's the whole idea, and it's also the weak spot. Fast growth is exactly what a fake project tries to manufacture. A few hundred cheap stars on a brand-new repo can push it into the top 10 of a niche topic.&lt;/p&gt;

&lt;p&gt;So every repo goes through filters before Upstar shows it anywhere: rules, a small classifier I trained on past decisions, and LLM judges that read the README and the files. Since October, they have blocked more than 3,800 repos.&lt;/p&gt;

&lt;h2&gt;
  
  
  8 stars, then 17 stars
&lt;/h2&gt;

&lt;p&gt;Upstar keeps a day-by-day star count for every repo it ranks. When I lined them up this week, one pattern jumped out.&lt;/p&gt;

&lt;p&gt;40 repos, owned by 40 different accounts with zero followers, all received exactly 8 stars on November 14. Three days later, on November 17, all 40 received exactly 17 stars. 25 of them had also received 3 to 5 stars on November 12. 38 of the 40 accounts were created from October 27 onward, most of them in batches.&lt;/p&gt;

&lt;p&gt;Different owners, different repos, identical deliveries on the same days. That's not 40 projects getting lucky. That's one seller filling one order.&lt;/p&gt;

&lt;p&gt;One caveat, because I almost got it wrong. Seven real projects also gained exactly 8 and 17 stars on those days, by pure chance: audiobookshelf, newsnow, mlx-audio and others, between 1,000 and 22,500 stars. Their curves are smooth, with stars coming in every day. The numbers alone prove nothing. The numbers on dozens of brand-new, empty accounts do.&lt;/p&gt;

&lt;h2&gt;
  
  
  Real project names, no real code
&lt;/h2&gt;

&lt;p&gt;Some of these repos borrow the name and description of a real open-source project and add a prefix. One was called &lt;code&gt;resp-expandModeIntegerScaler&lt;/code&gt;, a copy of a real Godot addon. Another copied a retro game project the same way.&lt;/p&gt;

&lt;p&gt;The code isn't there. Each of these two repos holds just two files: an HTML page and a &lt;code&gt;CONTRIBUTING.md&lt;/code&gt;. The README is empty, so the bait sits in the contributing file instead:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;a big button to a download on an outside site;&lt;/li&gt;
&lt;li&gt;a password for the archive;&lt;/li&gt;
&lt;li&gt;and the line that should make anyone stop: "Run as Administrator".&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The prefix "resp" shows up in the repo names of 136 young accounts. Another 66 accounts were created on October 28 within 35 minutes, many of them with repo names built on the same few words.&lt;/p&gt;

&lt;h2&gt;
  
  
  A cheat network built in 47 seconds
&lt;/h2&gt;

&lt;p&gt;The second campaign sells cheats for an online game called Rob Run. On October 14, six GitHub accounts were created between 13:23 and 13:40 UTC, five of them within 47 seconds. Each account created its repo about two minutes later.&lt;/p&gt;

&lt;p&gt;Every README tells you to run &lt;code&gt;robrun-hk.exe&lt;/code&gt;. That file isn't in the repo. It comes from a zip uploaded by hand to the releases, and the code in the repo doesn't build it.&lt;/p&gt;

&lt;p&gt;They were careful in small ways:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;each zip has a different size, from 1.65 to 2.92 MB, so they don't look like copies of one file;&lt;/li&gt;
&lt;li&gt;the repos hold 9 to 16 real-looking code files, so they don't look empty;&lt;/li&gt;
&lt;li&gt;the forks were bought too: 41 of one repo's 80 forks came from accounts all created on November 7.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;One copy of the same cheat went further. Its button ran a PowerShell one-liner that fetched a script from a GitHub Pages site, and the script decoded a payload with base64 and XOR. I haven't analyzed what it installs.&lt;/p&gt;

&lt;h2&gt;
  
  
  What my filters missed
&lt;/h2&gt;

&lt;p&gt;I'd rather tell you where Upstar failed than pretend it didn't.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;The executable rule only looked at near-empty repos.&lt;/strong&gt; I flagged "run this .exe" only when a repo had almost no code. The Rob Run repos had enough files to slip under it. Now a game cheat that asks you to run a program the repo doesn't contain is blocked, however much code sits next to it.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;The shared-file rule wanted identical files.&lt;/strong&gt; Six zips of six different sizes passed as six different files.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;The judges were too polite about cheats.&lt;/strong&gt; My policy kept online-game cheats out of the spotlight without blocking them. The judges saw real code and a zip hosted on GitHub, and they kept them. That's fixed in the policy now.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Old approvals never got a second look.&lt;/strong&gt; Two of the "resp" repos were approved on November 6. At the time, the judges didn't read the &lt;code&gt;CONTRIBUTING.md&lt;/code&gt; when the README was empty, so they never saw the bait. Approvals made before that change are now re-read.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;The classifier knew, but nobody asked it.&lt;/strong&gt; My classifier scored the Rob Run repos up to 0.998 out of 1. But it could only confirm a block that the first judge proposed. When the judge said "keep", the score went nowhere. Now any repo it scores at 0.9 or more goes to a stronger model for a decision.&lt;/p&gt;

&lt;p&gt;What did work: since November 18, that classifier has blocked 268 lures on its own, and none of those blocks has been reversed so far.&lt;/p&gt;

&lt;h2&gt;
  
  
  How to spot one yourself
&lt;/h2&gt;

&lt;p&gt;None of these signs is proof on its own. Two or three together should make you close the tab.&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;The account is as young as the repo.&lt;/strong&gt; Created the same day, zero followers, two or three commits.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The program isn't built from the code.&lt;/strong&gt; The README tells you to run an .exe or a zip from the releases, but nothing in the repo produces it.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The download lives somewhere else.&lt;/strong&gt; A button to Dropbox, a file host or an unknown site, often with an archive password.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;It asks for too much.&lt;/strong&gt; "Run as Administrator", "disable your antivirus", or a one-line PowerShell command to paste.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The README is empty and the bait is elsewhere.&lt;/strong&gt; Check &lt;code&gt;CONTRIBUTING.md&lt;/code&gt; too.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The name belongs to someone else.&lt;/strong&gt; A real project's name with a prefix or suffix added, and none of its code.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The stars came in steps.&lt;/strong&gt; A tool like a star-history chart shows big jumps on a few days instead of a steady curve.&lt;/li&gt;
&lt;/ol&gt;

&lt;h2&gt;
  
  
  Where this leaves me
&lt;/h2&gt;

&lt;p&gt;A ranking that rewards growth will always attract people who sell growth. I don't think I'll ever be "done" with this, any more than a spam filter is done with spam. Every week the tricks change a little: a different zip size, a few more code files, a real project's name.&lt;/p&gt;

&lt;p&gt;What I can do is make each trick cost more, and show the evidence. Upstar's rankings are free, with no account: &lt;a href="https://upstar.dev" rel="noopener noreferrer"&gt;upstar.dev&lt;/a&gt;. If you follow a topic there and see something that looks off, tell me. And if you've seen these star deliveries or this "resp" family somewhere else, I'd love to compare notes in the comments.&lt;/p&gt;

</description>
      <category>security</category>
      <category>github</category>
      <category>opensource</category>
      <category>webdev</category>
    </item>
    <item>
      <title>I found 560 malware repos while building a GitHub trending site. GitHub removed them in a day</title>
      <dc:creator>MAT Armel</dc:creator>
      <pubDate>Mon, 02 Nov 2026 13:48:24 +0000</pubDate>
      <link>https://dev.to/alone225/i-found-560-malware-repos-while-building-a-github-trending-site-github-removed-them-in-a-day-13k5</link>
      <guid>https://dev.to/alone225/i-found-560-malware-repos-while-building-a-github-trending-site-github-removed-them-in-a-day-13k5</guid>
      <description>&lt;p&gt;I build &lt;a href="https://upstar.dev" rel="noopener noreferrer"&gt;Upstar&lt;/a&gt;, a site that ranks GitHub repositories by how fast they gain stars within each topic. The idea is simple: a small project that suddenly takes off should be able to outrank a giant that grows slowly.&lt;/p&gt;

&lt;p&gt;Ranking by growth has an obvious weakness, though. Fast growth is exactly what fake projects try to manufacture. Last weekend I found out how far that goes.&lt;/p&gt;

&lt;h2&gt;
  
  
  It started with 17 "installers"
&lt;/h2&gt;

&lt;p&gt;Upstar runs a daily data-quality check. On Saturday it flagged a handful of repos that had no business being in a ranking: "pre-activated" versions of KMS activators, Adobe After Effects, Cubase, Microsoft Project, a MetaTrader "bot". They were sitting quietly in the Python rankings, around 300 stars each.&lt;/p&gt;

&lt;p&gt;I asked Claude Code to look for the same pattern across every README in Upstar's database. 17 became 309. By the time I wrote the report, the count was &lt;strong&gt;479 repositories on 478 accounts&lt;/strong&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  What they looked like
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Throwaway accounts&lt;/strong&gt;: random word-salad usernames, no followers, one or two commits.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Waves&lt;/strong&gt;: 132 repositories created within about two hours on October 29, and 322 more over the next three days.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Bought stars&lt;/strong&gt;: 450 of the 479 gained between 80 and 400 stars within days of creation.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The bait&lt;/strong&gt;: free "pre-activated" paid software (Adobe, Autodesk, Office, FL Studio, CCleaner, phone unlocking tools), game cheats, crypto "bots".&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  How the trap worked
&lt;/h2&gt;

&lt;p&gt;Every README asked the reader to do two things:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;paste a PowerShell command that downloads a script from a domain outside GitHub and runs it in memory;&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;turn off Windows Security's real-time protection first.&lt;/strong&gt;&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;In about half of them, the domain name was split into pieces and glued back together inside the command, so it never appeared as one string. I'm not reproducing any of the commands here. The scripts were served from four domains, the busiest being &lt;code&gt;git-software[.]cc&lt;/code&gt; and &lt;code&gt;shell-activation[.]top&lt;/code&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  A second, quieter wave
&lt;/h2&gt;

&lt;p&gt;While checking, we found 81 more repos built differently. Each contained a tiny placeholder program, about 1.6 KB, that GitHub Actions compiled into a 67 MB executable, complete with checksums and an MIT license, so it looked legitimate. The README, meanwhile, sent readers to a MediaFire folder for the "real" installer. All 70 accounts of the first batch were created on the same day.&lt;/p&gt;

&lt;h2&gt;
  
  
  How they got into my rankings
&lt;/h2&gt;

&lt;p&gt;This is the uncomfortable part. Upstar only scanned READMEs for bait above 500 stars, and these repos were designed to stay around 300. At one point, &lt;strong&gt;31 of the 200 fastest-rising projects on Upstar belonged to the campaign&lt;/strong&gt;, and 220 of the repos were visible on topic pages. Coordinated stars are cheap, and a growth ranking rewards them.&lt;/p&gt;

&lt;h2&gt;
  
  
  Reporting it
&lt;/h2&gt;

&lt;p&gt;On November 1, I sent one report to GitHub Trust &amp;amp; Safety, with the full list in a secret gist.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;About four hours later: "Our team is currently investigating."&lt;/li&gt;
&lt;li&gt;The next morning, they asked for the full list of accounts, and I sent all 559.&lt;/li&gt;
&lt;li&gt;A few hours after that, the review was concluded, and every account from both reports was gone.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;From one report to 559 removed accounts in under 24 hours. That's a genuinely good response.&lt;/p&gt;

&lt;h2&gt;
  
  
  What I changed on Upstar
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;README scanning for bait now runs on every repo, whatever its star count.&lt;/li&gt;
&lt;li&gt;A README that runs a remote script &lt;strong&gt;and&lt;/strong&gt; asks you to disable your antivirus is treated as a lure, full stop. No legitimate project asks for both.&lt;/li&gt;
&lt;li&gt;Links to anonymous file hosts count only when they come with other red flags, such as a brand-new account or a handful of commits, so real projects that host big files elsewhere aren't caught.&lt;/li&gt;
&lt;li&gt;A block now follows the repository's ID, so renaming or transferring a repo doesn't get it back in.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  How to spot one yourself
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;A paid product offered "free", "pre-activated" or "cracked".&lt;/li&gt;
&lt;li&gt;The download lives outside GitHub: a MediaFire link, a random domain, a one-line PowerShell installer.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Any instruction to turn off your antivirus.&lt;/strong&gt; That alone is enough to walk away.&lt;/li&gt;
&lt;li&gt;A brand-new account, no followers, one or two commits, and hundreds of stars in a few days.&lt;/li&gt;
&lt;li&gt;A "release" binary that the repository's own code obviously can't produce.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;If something trending on GitHub ticks two of these boxes, report it. It takes five minutes, and GitHub does act.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Upstar ranks GitHub repos by star velocity within each topic, and now screens out this kind of bait automatically: &lt;a href="https://upstar.dev" rel="noopener noreferrer"&gt;upstar.dev&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;

</description>
      <category>security</category>
      <category>github</category>
      <category>opensource</category>
      <category>showdev</category>
    </item>
    <item>
      <title>My GitHub trending site lost 84% of its Google traffic in one day. Here's what I got wrong.</title>
      <dc:creator>MAT Armel</dc:creator>
      <pubDate>Thu, 29 Oct 2026 11:12:39 +0000</pubDate>
      <link>https://dev.to/alone225/my-github-trending-site-lost-84-of-its-google-traffic-in-one-day-heres-what-i-got-wrong-53he</link>
      <guid>https://dev.to/alone225/my-github-trending-site-lost-84-of-its-google-traffic-in-one-day-heres-what-i-got-wrong-53he</guid>
      <description>&lt;h2&gt;
  
  
  Quick context
&lt;/h2&gt;

&lt;p&gt;I'm a solo dev in Abidjan. On October 13 I launched &lt;a href="https://upstar.dev" rel="noopener noreferrer"&gt;Upstar&lt;/a&gt;, a site that ranks the GitHub repos taking off inside each topic: self-hosted, Rust, local LLMs, Android, agent skills, 69 topics so far.&lt;/p&gt;

&lt;p&gt;The idea is simple. GitHub Trending is one big list where the same giants win every day. On Upstar, every topic is its own little league, and growth is weighed against size. So a 300-star project that doubles in a week can beat a giant that gains 2,000.&lt;/p&gt;

&lt;p&gt;I built it with Claude Code. That detail matters later.&lt;/p&gt;

&lt;h2&gt;
  
  
  Ten good days
&lt;/h2&gt;

&lt;p&gt;Google picked it up faster than I expected. 14 clicks on the first full day, 51 the next, 159 on October 19. I refreshed Search Console way too often.&lt;/p&gt;

&lt;p&gt;By October 22, Upstar was showing up about 6,450 times a day in Google, with 147 clicks. About 20 people had signed up for the Monday digest. Most of the traffic came from people typing a project's exact name, like "macviewer github", and finding my page about it in the top five.&lt;/p&gt;

&lt;h2&gt;
  
  
  Then October 23
&lt;/h2&gt;

&lt;p&gt;Impressions went from 6,450 to 1,031 in one day. Clicks from 147 to 54. No manual action, no server errors, and I hadn't even deployed anything that day.&lt;/p&gt;

&lt;p&gt;What confused me was the average position. It barely moved, around 6.7. It took me a while to get it: Google wasn't pushing my pages down, it had just stopped showing most of them. The few it still showed ranked fine, so the average looked healthy.&lt;/p&gt;

&lt;p&gt;So if you ever see impressions collapse while the average position holds, look at indexing first.&lt;/p&gt;

&lt;h2&gt;
  
  
  What Google actually saw
&lt;/h2&gt;

&lt;p&gt;The day before the drop, Googlebot fetched about 16,500 of my pages, way more than on any day before. And the day before that, I had shipped a feature that added a README excerpt to every repo page. Bad timing.&lt;/p&gt;

&lt;p&gt;So Google read thousands of pages built from the same template: repo name, the GitHub description, a chunk of README, a few numbers. An audit tool said only 32% of my content was unique. My own count was worse, with a median of 22% of words that actually belonged to each page.&lt;/p&gt;

&lt;p&gt;And I had a French version of almost every page. That's about 39,000 URLs that mostly repeated what GitHub already says.&lt;/p&gt;

&lt;p&gt;Google's &lt;a href="https://developers.google.com/search/docs/essentials/spam-policies" rel="noopener noreferrer"&gt;spam policies&lt;/a&gt; have a name for this: scaled content abuse. Lots of pages, little value, no matter how they're made. The same page explains why my good pages went down too: Google tends to assume a page is about as good as the rest of the site.&lt;/p&gt;

&lt;p&gt;The annoying part is that Claude knew about these policies. It just didn't bring them up, and I never asked.&lt;/p&gt;

&lt;h2&gt;
  
  
  Fix #1, and the mistake inside it
&lt;/h2&gt;

&lt;p&gt;First I took about 82% of the pages out of the index. They stayed online for visitors, just marked noindex. A repo page stayed indexed only if the repo had made a topic's top 50 at least once in 30 days, or had already brought a visitor from Google. That cut 39,232 URLs down to 7,151, while keeping most of the pages that actually got clicks.&lt;/p&gt;

&lt;p&gt;I felt good about it for about three days.&lt;/p&gt;

&lt;p&gt;Then I opened a fresh Claude Code session with no history and asked it to look at my SEO cold. It simulated the rule on real data and found the problem. New repos enter a top 50 every day, so the index was growing back by about 1,500 pages per language every week. By mid-November I'd have been right back where I started.&lt;/p&gt;

&lt;p&gt;Now the rule is stricter: top 10 of a topic in the last 7 days, or a past Google click. That's about 1,200 English repo pages and 900 French ones, and it stays stable. There's a hard cap per topic, a test that fails if the cap is broken, and an email alert if the index grows more than 10% in a week.&lt;/p&gt;

&lt;h2&gt;
  
  
  The other changes
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Every repo page now starts with why the project matters this week, which GitHub doesn't tell you. For example: "gaining 3 times more stars than usual, Upstar spotted it 9 days ago at 336 stars." Then a 30-day chart with the day it was spotted, its rank in each topic and a small health check. The README excerpt is gone.&lt;/li&gt;
&lt;li&gt;I redesigned the whole site so you get it in five seconds, with one line explaining how the ranking works.&lt;/li&gt;
&lt;li&gt;I stopped flooding Bing. I was sending about 6,500 URLs a day to IndexNow, mostly pages where only the star count had changed. Now it's around 800, and only real changes.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  The number I didn't want to see
&lt;/h2&gt;

&lt;p&gt;Search Console's own reports lag by several days, so I wired up the URL Inspection API to ask Google about every URL directly.&lt;/p&gt;

&lt;p&gt;Out of 1,375 English URLs, 173 are indexed. Google has never even heard of 452 of them. It knows about 227 but hasn't crawled them yet. And it crawled 523 and decided not to keep them.&lt;/p&gt;

&lt;p&gt;It hurt to read, but it's the first honest number I've had. Half of my pages haven't been rejected, they just haven't been looked at yet. The 523 are the real problem, and that's what all the changes above are about. Now I can watch that number every week instead of guessing.&lt;/p&gt;

&lt;h2&gt;
  
  
  What I'd tell another indie dev
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Don't publish thousands of pages built from someone else's data. Every page you want indexed should say something no other page says.&lt;/li&gt;
&lt;li&gt;Look at why pages aren't indexed. "Crawled, currently not indexed" means Google judged them. "Unknown" or "discovered" usually just means it hasn't visited yet.&lt;/li&gt;
&lt;li&gt;Stop checking the 24-hour view in Search Console. The last couple of days are always incomplete, and it'll ruin your mood for nothing. I learned that one the hard way.&lt;/li&gt;
&lt;li&gt;Before you ship anything that touches thousands of pages, ask your AI what could go wrong with search, security or legal. It won't always tell you on its own.&lt;/li&gt;
&lt;li&gt;Get a second pair of eyes. A fresh session with no context caught a mistake I had walked right past.&lt;/li&gt;
&lt;li&gt;Don't panic-edit. Google needs weeks to re-evaluate a site.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Where things are now
&lt;/h2&gt;

&lt;p&gt;Too early to say if Google comes back. Meanwhile Upstar keeps running: rankings update every hour, and the Monday digest goes out to everyone who signed up.&lt;/p&gt;

&lt;p&gt;If you have five seconds, open &lt;a href="https://upstar.dev" rel="noopener noreferrer"&gt;upstar.dev&lt;/a&gt; and tell me what you think it does. And if you pick a topic you know well, does the ranking make sense to you? Honest answers welcome, harsh ones included.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Armel, solo dev in Abidjan, Côte d'Ivoire. I wrote this with some help from Claude for the English; the numbers and the mistakes are mine.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>seo</category>
      <category>webdev</category>
      <category>showdev</category>
      <category>sideprojects</category>
    </item>
  </channel>
</rss>
