GitHub Outage Map

Community Discussion

Tips? Frustrations? Share them here. Useful comments include a description of the problem, city and postal code.

Beware of "support numbers" or "recovery" accounts that might be posted below. Make sure to report and downvote those comments. Avoid posting your personal information.

GitHub Issues Reports

Latest outage, problems and issue reports in social media:

• 
  Brayy (@0Drayne) reported 54 seconds ago

  Meet ShinyHunters This crew is mostly in their 20s, but they’ve been tearing it up since 2020. We’re talking Tokopedia, Microsoft, AT&T you name it. They’re basically making a killing off simple mistakes. Most of the time, they don’t even need high-tech tools. They just look for things that slip through the cracks, like API keys left in public GitHub repos or AWS buckets that aren't locked down. It’s wild how often big companies still mess this up.

• 
  Mehmet Gonullu 🎙️🎯🚀🔁 (@m_gonullu) reported 56 seconds ago

  $1T came off AI and chip names Friday. Worst Nasdaq day since April 2025, on one question: can the model layer earn its capex. That is the consensus story. The more useful signal was one layer over. Same week: a single GitHub issue could hijack repos running Claude Code via prompt injection. A related flaw in February pushed a rogue package onto ~4,000 dev machines in 8 hours. Agents in production are now attack paths. Microsoft's answer at Build: every agent isolated, carrying its own identity, every action logged. The moat in this phase is not the benchmark. It is giving an agent real tools without giving away the keys.

• 
  Searxly (@Searxly) reported 1 minute ago

  Plans for today: - Add more tabs to the website, redesign, fix bugs on mobile. - Redesign entirely search results in Searxly. - Implement the Wallet feature inside of Searxly, make it work - Publish all changes today to the GitHub repository.

• 
  Tejas Vachhani (@imtejasvachhani) reported 3 minutes ago

  GitHub Copilot (AI + Momentum) The physics: Momentum p = m·v — mass (substance of your skill) times velocity (speed of execution). AI acts as a force multiplier on v, but cannot supply m. Application: A developer's mass is their understanding of system architecture, problem logic, and code quality. Velocity is how fast they type and debug. Copilot eliminates the high-friction parts of velocity: boilerplate code, syntax lookup, repetitive patterns. The developer stays in flow state longer, so their velocity increases dramatically. But if a junior dev with no mass (no architectural understanding) uses Copilot to ship code at high velocity, the result is a fragile, buggy system — fast garbage. The winning formula: solid senior developers amplify their existing mass with AI velocity, building momentum that's incredibly hard to stop.

• 
  Sipping on AI (@sipping_on_ai) reported 4 minutes ago

  GitHub Copilot has a desktop app now. The useful test: Can it take 5 stale issues, make reviewable PRs, and explain every file it touched? Try it on backlog dust first. Not the checkout flow.

• 
  Trinks | Making Alerith (@xTrinks) reported 5 minutes ago

  @0xanmol How do I commit my code to GitHub can u help I don’t know what to do the command “github sendcodecommit allcode add comment no claude” is not working ser

• 
  jack gamrot (@jgamrot) reported 5 minutes ago

  @HudsonByrd77 @JessePeplinski Sorry, I wasn’t very clear. I get the curl part and I can see the technical flow now. I think what threw me is that I was expecting the landing page/demo to give me the product aha moment, so I assumed I’d missed something when it didn’t click. I was looking for something more concrete like: Claude hits an error, searches docs and GitHub, tries a few fixes and burns 20k tokens. vs Claude queries Olleey, gets back the fix that already worked for everyone else and moves on. That would have made the value proposition click for me immediately, and is probably why I thought the demo wasn’t working.

• 
  Bijan (@Beethoven779) reported 7 minutes ago

  @ryanvogel opencode has a lot of potential. I use it everyday, but I am not happy with it to be honest. There is UX issues there, I mention in X and in github issues, they either get ignored or prs closed because certain time has passed and they did not have time to review or...

• 
  Gary Bernhardt (@garybernhardt) reported 11 minutes ago

  2000: You're free to submit a patch to the mailing list! 2010: You're free to open a GitHub issue! 2020: You're free to submit a PR to the GitHub repo, but issues are closed! 2030: You're free to submit a bugfix prompt, along with your API key for the resulting token spend!

• 
  Kolo Not Muani (@koloz_IV) reported 12 minutes ago

  @akashdeepsb @MrPunyapal @github on a request header lol, took me 2days prompting lovable to locate this and fix it it's terrible not know all corners of your code

• 
  bob_irl (@bobIRL__) reported 20 minutes ago

  The best AI products never mention AI. Grammarly just fixes your typos. No "POWERED BY ML!" banners. GitHub Copilot completes your code. No neural architecture explanations. Nobody cares about your tech stack. They care about their problems getting solved.

• 
  Matt Hulme (@MattHProgrammer) reported 22 minutes ago

  @gagansaluja08 Really depends on the work I'm doing, but here are a couple examples: • Every 5 minutes, check GitHub Actions. If a build fails, attempt to fix the issue and open a commit • Check every 15 minutes whether DNS propagation is complete and SSL certificates are issuing correctly

• 
  Tithi (@SaptamiBis76657) reported 28 minutes ago

  I was trying to sign up for Github educational plan but they r asking for clg email od rn I have email I'd can't login through Outlook. If I only provide College id will it be considered? @github

• 
  Byte | yoursaudit (@yoursbyte) reported 30 minutes ago

  @F0rkedNft @immunefi If they address the issue into github then it is ok but i havent found when you see into immunefi they even dont disclose which protocol even give the bounty

• 
  nicole (@nicolekcha) reported 30 minutes ago

  Calling this an “outage, not a bug at the inference layer” is a sleight of hand here. It’s still a bug *somewhere* even if not *at the inference layer*. And it’s one of the worst classes of bugs. If an outage can cause your responses to go to the wrong requester, across tenant boundaries, it’s still an *extremely bad* bug. Even if it is due to a bug in a dependency, like a load balancer — you still passed that bug down to your users. You can root cause it and do a post mortem, but it still ultimately is your responsibility. (Like the Vercel data breach through an AI tool an employee used or the supply chain attack on TanStack through a Github Actions bug). Many users reading will ride along with the wording “outage, not a bug in our inference” and not realize “still a bug somewhere in the stack”, taking comfort from a technical distinction that doesn’t earn it. And if that’s what was intended here, that’s a very bad look, and a PR red flag.