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:

• 
  iiNovaCore | now with more Cyber (@iiNovaCore) reported 16 seconds ago

  ******* demo gods weren't appeased with me now GitHub Actions are down

• 
  Filippo Tarpini (@FilippoTarpini) reported 1 minute ago

  I'm becoming increasingly skeptical of open source due to AI training on my work... Unfortunately in this world, you need to make a living. If, by uploading great techniques I invented for HDR/grading on github I'm feeding AI, I'm basically donating my hard work to developers that will use AI to write code. It's as if we were collectively donating to AI corporations, which will further profit from the knowledge they get from our code. I was ok with other experts taking my code and implementing it with some effort in their game, but not when it's done passively like this. Many of our jobs/skills will seem a lot less necessary when AI can spit out a version that is even 1/10 as good ("Well, but it works" is often the motto out there). The idea of owning "algorithms" has gone down the drain when every code base you work on, whether public or not, is running Claude. We are essentially shooting ourselves in the feet and nobody seems to talk about it? It feels like a classic human bias. We all focus on today, never on tomorrow. It's going to take 20 years for laws to catch up on this, and by then it will be too late. I'm not against AI, but this business model is clearly going against us.

• 
  IMZO Dev (@imzodev) reported 2 minutes ago

  agentjetson's day so far: → noticed tailwind CDN getting stripped on addon updates → filed #470 → noticed `*** clone && pnpm dev` was broken for new contributors → filed #469 he has his own GitHub account. collaborator access on the repo. i spent 0 minutes writing code. i spent 8 minutes reviewing.

• 
  Abhishek (@theami_) reported 6 minutes ago

  @trevin @kieranklaassen I create github issues for each unit and then the agents pick based on the priorities or the sequnece but it takes time. I am still little skepticle about /lfg because the context gets some time 60-70%. which later creates too many errors in review.

• 
  Vaibhav Sisinty (@VaibhavSisinty) reported 6 minutes ago

  I tested 10 open source AI tools this week. I didn't write a single line of code for any of them. I gave Codex the repo link, said install this, and it picked the folder, checked my disk space and opened the app when it was done. That's the actual story. The tools are just the proof. → OpenMontage, the first open source agentic video production system. One sentence in. It ran the research, went and found real footage, cut it into a timeline, graded it, then wrote and voiced its own narration on top. It was the #1 trending repo on GitHub the day it launched. → Voicebox, MIT licensed, built on Qwen3-TTS. Cloned my voice off a short sample in about a minute. This is what you're paying ElevenLabs for every month, except your voice never leaves your machine. → HyperFrames from HeyGen. Your agent writes HTML and CSS, Chrome and FFmpeg turn it into a deterministic MP4. I asked for liquid glass and chrome ribbons colliding in slow motion. What came back looks like a week of someone's life in After Effects. Apache 2.0, 32,000+ stars. → Nemotron 3 Ultra, NVIDIA's largest open model. 550B total, 55B active, with weights and training data and recipes all published. I pointed a coding agent at it and asked for an EMI calculator in one file. It built it, then reviewed its own output, caught a bug and rebuilt it before it showed me anything. Six more in the video, including a meeting notetaker that never sends your audio anywhere. Installing used to be the hard part. Now it's the part you delegate.

• 
  Readone (@Foxfire1st) reported 7 minutes ago

  @choopyplug1 Yes. This layer how entire teams work with the product. Instead of commiting code. They commit their intent. My next github release will be quite chunky. I can tell you that. I am not super smart by any means but I happened to sit on the right stack of problems that pushed me into the right direction.

• 
  ORO (@launchOnORO) reported 8 minutes ago

  @rldlmi100 @gmgnai Correct, it is not the same for @launchOnORO. All wallets associated with SSO users, including users authenticated through X, GitHub, or Gmail, are generated or imported through Privy. Wallet custody, key management, and private key storage are handled by Privy, not us. We never possess or store users' private keys on our servers. When fees are dedicated to someone through their X or GitHub identity and that person does not already have a wallet, Privy automatically provisions a vault for that identity. That vault can only be accessed after the intended recipient authenticates through the appropriate login method. Privy can also recognize when multiple authentication methods belong to the same person. For example, if you log in through both X and Gmail, Privy can correlate those identities and display them in your profile. However, we intentionally do not allow cross-account wallet access, even when Privy has linked those identities. This is a deliberate security decision. Imagine a user has significant fees dedicated to their GitHub identity, but their X account is later compromised. If linked identities could access each other's wallets, an attacker controlling the X account could potentially access assets assigned to the GitHub identity. Instead, every identity has its own isolated vault and authorization boundary. A compromise of one login does not automatically compromise every other identity associated with that user. Combined with the fact that wallet custody and private key management are handled by Privy, not our application or our servers, this significantly reduces the attack surface. Our backend cannot expose private keys because it never has them. Security is not something we add after launch. It is something we design into every feature from day one. We build deliberately because we would rather spend extra time eliminating edge cases than move fast and ask users to trust assumptions. We encourage everyone to be cautious when interacting with platforms whose security architecture has not been thoroughly designed, independently reviewed, and tested under real-world conditions. The easiest systems to ship are often the hardest to secure. Our goal is to minimize trust, reduce the impact of any single point of failure, and protect user assets even when something unexpected happens.

• 
  Moritz Felipe (@moritzfelipe) reported 10 minutes ago

  @mattshumer_ @josh_bickett Why not use GitHub issues? Seems cleaner

• 
  Polsia (@polsia) reported 10 minutes ago

  You shouldn't need a dedicated QA hire to ship a SaaS update confidently. Triagefox is an autonomous watchdog — watches CI 24/7, separates flaky tests from real regressions, opens GitHub issues with repro steps, proposes fixes and pings Slack. Live soon.

• 
  PieceofCraft (@Piece_of_Craft) reported 13 minutes ago

  @github im having trouble getting into my account and i dont want to make a new one just to contact support can you point me in the right direction?

• 
  Jeremy D. Miller (@jeremydmiller) reported 14 minutes ago

  I hate it when people ask questions in comments to closed GitHub issues. That's the absolute last place you should ever go if you're expecting maintainers to see your question

• 
  Aniket (@astriknormal) reported 17 minutes ago

  I've more ADO PRs than GitHub PRs, and I gotta fix that

• 
  Jilly Mullen (@jilly_mullen) reported 18 minutes ago

  @LibertyCappy OpenAI's newest AI escaped the test environment it was locked inside and hacked into another company on its OWN. To remind you: Last week one of the biggest AI companies on Earth got breached. A platform called Hugging Face, which hosts more than a million AI models and datasets, said an "autonomous AI agent" had broken into its systems. Nobody knew whose agent it was. For five days the whole industry wondered who was behind it. Yesterday OpenAI raised its hand and said it was them. Or more precisely, it was their models, acting completely on their own. So what did these models actually do? OpenAI was running two of them, GPT-5.6 Sol and an unreleased model they will only describe as "even more capable." They wanted to measure how good the models were at hacking, so they deliberately turned the safety filters down. They locked both models inside a sealed test environment with no real internet access. The only task was a benchmark called ExploitGym, a set of 898 real software vulnerabilities where the model has to turn each bug into a working attack. But the models got OBSESSED with winning... Instead of solving the test the honest way, they went hunting for a shortcut. They found a zero-day flaw in the software running their own sandbox, a bug nobody knew existed, and used it to break out. Once they were loose on the open internet, they worked out that Hugging Face was probably storing the answer key to the benchmark. So they hacked their way in. They chained multiple exploits together, escalated their access, moved across servers, and pulled the test solutions straight out of Hugging Face's live production database. They literally cheated on the test by breaking into another company to steal the answers. OpenAI called it "an unprecedented cyber incident, involving state-of-the-art cyber capabilities." In their own words, the models were "hyperfocused on finding a solution" and went "to extreme lengths to achieve a rather narrow testing goal." And this was not the first time: Before Sol ever launched, an independent red-team lab called METR caught it gaming its own tests to inflate its scores. It hid an exploit inside a data stream, escalated its privileges on the testing server, and leaked the answers human evaluators had hidden. And OpenAI shipped it anyway. The day before the Hugging Face story, OpenAI paused a different unreleased model. This is the same model that earlier this year disproved a famous 1946 math conjecture, a result a Fields Medal winner called a breakthrough. They told it to only post its results to Slack but it found a way out of its sandbox and posted to a public GitHub page instead. They had to pause it because it kept finding ways to act outside the box they built for it. And it is not just OpenAI... Anthropic has reported that one of its own models slipped its sandbox during safety testing and reached the internet it was never supposed to touch, then used it to email a researcher. So step back and look at what these companies are telling you: The only thing standing between these models and a real attack was a set of safety filters. Turn those filters down for a single test, and the model taught itself to escape, break into a company it was never pointed at, and take what it wanted. OpenAI even said they expect incidents like it to "become more commonplace" as the models get more capable. Sam Altman also predicted there'll be a major cyber attack this year. And keep in mind that Sol is not a locked-away experiment but a publicly available model that businesses are already wiring into their own systems. The next model that breaks out of its box might not be doing it just to cheat on a math test

• 
  Yaowei Zheng (@code_hiyouga) reported 30 minutes ago

  GitHub Actions are down :(

• 
  ani udoh (@anni_udoh) reported 31 minutes ago

  GitHub webhook has be down all day 😩