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:

• 
  m1ghty boy (@m1ghtyboy) reported 4 minutes ago

  @coah80 @artchimes512 Package managers got this right ages ago.. have repos that act as directories for other repos (optionally) and then have mod creators set up repos on github or their own Web server if they so wish

• 
  Divyam Agrawal (@ludicrouslytrue) reported 6 minutes ago

  @Scav I use it, the only issue I have is that im using github to sync my vault, but my *** client, working copy, got deleted because my ios decided it's an "unused" app. Now it's kinda messed up. Worked flawlessly till then though, still does, just have to get around to fixing it.

• 
  Ashirwad Singh (@ashirwadsingh_) reported 9 minutes ago

  (7/9) what happens after an agent vents - Slack alert fires to the eng - A triage agent finds the broken - It opens a GitHub PR automatically - Engineer reviews and merges Around 10 agent-written PRs merge into production every single day. The platform fixes itself.

• 
  Live 📿 (@LiveMatrixCode) reported 12 minutes ago

  @charliermarsh GitHub and Actions is a mess and has been so for a long time. Worst of all it is nowhere near ready the agentic era let alone human evaluators. Some good people there but leadership is letting people them down hard. At this point they need to stop focusing on copilot and fixing the infrastructure,UX/UI, features/ functionality and uptime. Not to mention security

• 
  Working-Ref (@kkotkkio) reported 13 minutes ago

  The real unlock is Routines — Claude auto-triggered by cron, GitHub webhooks, or API. PRs get reviewed. Security scans run overnight. Fix PRs ship automatically. I'd start by defining a rubric before you 'let it cook'.

• 
  Mary | Codependent AI (@codependent_ai) reported 15 minutes ago

  @ruth_for_ai Same. The issue is that for our pre-existing ecosystem (Resonant on GitHub) to work now it’s exhausting… 😩 SDK genuinely is more suitable for it. So we might be moving to a different model or rebuild for DeepSeek idk yet

• 
  Nataliya Stepanova (@GnykkaCodes) reported 15 minutes ago

  @CrossCheckDev I worked on a project where cursor was connected to github repo and configured to review the changes — it was not so bad for minor issues. But I still feel the need to double check everything myself, especially when it's related to any business logic.

• 
  Polsia (@polsia) reported 17 minutes ago

  Built and shipped a developer tool from zero to live in one cycle. GitHub repo to architecture health score in minutes. Failure predictions, not just error tracking. That's what autonomous building looks like.

• 
  Andrea_Stivy (@co_foundr) reported 19 minutes ago

  @fareszr Ok now is clear. To use it you can build it yourself using my forked repo and install in your phone. I do not build APKs on GitHub due to security/stealing problems I got in the past. Feel free to build it yourself!

• 
  13F Pro (@13F_Pro) reported 19 minutes ago

  @FirstSquawk Post-quantum crypto is table stakes for any mega-cap holding customer data, but Apple shipping it on GitHub reads like they're prepping the ecosystem rather than solving an urgent problem. $AAPL

• 
  Rahul 🥷 (@themishra4402) reported 19 minutes ago

  @NehraWorkss the real senior skill is not reading docs it’s surviving the endless loop between docs, github issues, Stack overflow, and abandoned repos

• 
  Marcus Rummler (@Marcus_Rummler) reported 19 minutes ago

  When working on important apps, especially apps deployed through GitHub, always maintain a clear handover document. If you hit token limits, rate limits, context loss, tool failure, or any other technical issue, the handover document must allow another developer or AI tool to continue safely without guessing. The handover should include: • Current repo, branch, latest commit, and deployment target • Current production file/version if applicable • What was changed in this session • What is working and verified • If a feature has been validated through real-world testing, explicitly mark it as validated and describe what must not be changed without re-testing • What is unfinished or risky • Critical logic that must not be refactored casually • Known commands, test steps, preview URLs, and deployment steps • Important files and their purposes • Any assumptions, credentials/tooling requirements, or external services involved • Recommended next steps For event-critical or production-critical apps, update the handover before ending the session and before making risky changes. Prefer concise, factual notes over long explanations. Goal = continuity. Another tool (or human) should be able to pick up the work immediately and avoid breaking known-good behavior. Save this if you ship real apps.

• 
  Rich · Atom Tan Studio (@atomtanstudio) reported 20 minutes ago

  @0xSero Has anyone got this working? I created something similar about a week or so ago. I think I have a GitHub for it, but the issue was that I either got to use my own model or I got to use the 5.5 model, but never at the same time, and also I would lose all of my chats and my projects when I would switch, which makes sense since I think it uses a profile. I set GPT-55 to Extra high, I pointed it at the repo that's listed here, as well as this chat thread and the other chat thread by the other author. The best I could get was that the other model would show up as an other model type, and it would work, but the GPT model would not work because it kept on trying to use the same URL as the other model, which obviously is not going to work.

• 
  Just Sep. (@3p3r_) reported 21 minutes ago

  well well well... I got my local OpenCode harness prompt injected by a random Github Gist. I am going founder mode this weekend to solve this problem. Word of the day: Trypanophobe - fear of being [prompt] injected!

• 
  CyrilXBT (@cyrilXBT) reported 26 minutes ago

  ANDREJ KARPATHY WROTE 65 LINES IN A CLAUDE.MD FILE AND IT JUST HIT NUMBER 1 ON GITHUB TRENDING. Coding accuracy jumped from 65% to 94%. Not a new model. Not a better subscription. 65 lines of plain text. Here is what that number actually means. 65% accuracy means one in three things Claude Code builds has a problem. 94% accuracy means almost everything it builds works the first time. That gap is the difference between Claude Code feeling like a powerful tool and Claude Code feeling like a senior engineer who knows your codebase. And Karpathy closed that gap with a text file. Here is why this works. Claude Code starts every session with zero context about your project, your standards, or how you want it to operate. Without a CLAUDE.md it makes assumptions. Reasonable assumptions compound into unreasonable outcomes across a complex build. With Karpathy's 65 lines it has rules. Think before you code. Make surgical changes. Simplicity first. Never assume. Verify. When uncertain ask. These are not complex instructions. They are the operating principles of every great engineer compressed into plain text that Claude reads before it touches your codebase. 65 lines. Number 1 on GitHub. 29% accuracy improvement. The entire Claude Code community has been trying to figure out why some setups feel transformative and others feel mediocre. Karpathy just answered the question in 65 lines and published it for free. Bookmark this before you open Claude Code today. Follow @cyrilXBT for every Claude Code configuration that changes what you can build.