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:

• 
  Aviinfinity (@aviinfinity) reported 2 minutes ago

  @msnofficial_on Open a github issue

• 
  Dominik Koch (@dominikkoch) reported 3 minutes ago

  Either t3connect is only for certain members or its broken on stable as well tested with google/github. Error only happens when you log in twice first login attempt does nothing second errors before this. cc: @jullerino @theo

• 
  Blaine Brown (@blizaine) reported 4 minutes ago

  @sudoaptupdater I'm looking into it. But it would help if you could either post a bug with console logs on GitHub or DM me the console errors.

• 
  Thomas Peitz (@tpeitz_dus) reported 5 minutes ago

  @cjav_dev Yeah so basically my workflow is posting whatsapp feature requests into github issues and the rest claude does :D - I would connect those two but the security layer is not solved (yet)

• 
  Oliver Prompts (@oliviscusAI) reported 5 minutes ago

  merge dev charges $65 per linked account a month past your first 10, that's $65k monthly at 1000 connections. open-connector does the same job for $0. there are two ways to run it: option 1: self-host > docker compose up --build > full control, sqlite storage, mcp, openapi, and a web console, all local option 2: cloudflare deploy > same gateway, workers runtime instead of docker, d1 for state, r2 for file transit > lighter to host, no server to babysit bonus: if you're blocked on oauth approval or shipping under a deadline, oomol runs the hosted version with a path to migrate back to self-hosted later. 800+ saas providers either way, github to bigquery to slack.

• 
  seeco (@seeconvm) reported 8 minutes ago

  11 llm evaluation methods AI engineers should know two metrics can rank the same models in opposite order. both can still be valid because they measure different things (word overlap vs meaning) 1 - reference-based (when ground truth exists) - BLEU - ROUGE - BERTScore 2 - judge-based (when no ground truth) - G-Eval - LLM-as-Judge - LLM juries 3 - human and deterministic - human eval - DAG 4 - for agents - trajectory accuracy - multi-turn eval 5 - before production - safety eval most of these are already in Opik (open-source, 20k+ stars on GitHub). it runs evals on traced production data with just a few lines of code metrics only show which case failed. the rest (finding the bad span, fixing the prompt or tool, re-running, checking regressions) is still mostly manual a detailed walkthrough with code shows how to automate this full cycle using Opik: diagnose the failed trace, apply the fix on the same input, and keep it in the eval set as a regression case

• 
  Arpit (@_xonoxc) reported 13 minutes ago

  @Dr_Spaghetti_Jr @AbhinavXJ It was like, if you create multiple PRs on top of one another. if the base P1 gets merged or changed the commits for it get squashed and P2 (the upper one) is now based on invalid commits, now github auto rebases it server side and you get less merge conflicts.

• 
  shmidt (@shmidtqq) reported 13 minutes ago

  13 SKILLS TURN ONE AI AGENT INTO A WHOLE DEPARTMENT. 23 MINUTES, ZERO CODE An assistant costs $60K a year. This is the same output for one evening and $7 a month. Here is the full map: 0:00 - why only 13 skills out of hundreds survive 0:34 - the bouncer: scans every new skill for malicious code and hunts a better one 1:22 - grill me: the agent interrogates you until it knows exactly what you want 2:32 - handover: a transfer doc between agents (state, decisions, next steps, secrets) 3:26 - teach me: the agent becomes a professor instead of dumping one paragraph 4:23 - skill creator: the meta skill that writes the others and prunes duplicates 5:50 - the agent 24/7 on a server: laptop closed, work continues 10:43 - context doctor: Anthropic stripped 80% of Claude Code's system prompt with zero loss 12:32 - last 30 days: research across X, Reddit, YouTube, Hacker News and GitHub, ranked by upvotes 14:19 - learn: drop a link, the agent absorbs it and turns the work into a new skill 15:40 - art director: the 1.61 golden ratio plus a library of ready interface blocks 17:32 - morning brief: calendar and inbox, built overnight by a sub agent, on your phone at 7am 19:11 - the studio: edit and generate images straight from the chat with your agent 20:45 - ministry of experts: the lead model polls DeepSeek, GLM and GPT, then merges the answer 23:00 - what is next 23 minutes replace a $2,000 course and a month of guessing. One person + an agent + 13 skills = a department that never sleeps. Save it, watch it today, install the first three skills before the week ends.

• 
  Azzle (@AzzleAI) reported 17 minutes ago

  @GitHubCommunity, we’d really appreciate your help in raising awareness about this. Both our profile and repo are showing a 404, despite our work being merged into known projects’ skill libraries like @bankrbot. We also didn’t receive any email from GitHub about any issues with our account.

• 
  Archonic (@Archonic2) reported 21 minutes ago

  @Alphons63 could open an issue on their Github

• 
  Oliver Servín (@oliverservinX) reported 23 minutes ago

  I've built my own blog more times than I can remember. Every time it went the same way. I'd be happy with it, publish a few posts, then drift away. I kept renewing the domain, but eventually I'd delete the VPS behind it. The address stayed alive, pointing at nothing. And I'm fine with that. An old blog is a snapshot of an older me: the things I cared about, the tools I used, the writer I was. When it doesn't fit anymore, letting it expire feels more honest than keeping a neglected archive. What bugged me was never the abandoning. It was that every time I had a new post ready, the blog was gone. I'd have to rebuild one from scratch before I could put a single word out. So this time, with a finished draft and nowhere to put it, I thought about Bearblog, a platform built on the idea that a blog needs almost no CSS and almost no layout: a header, a list of posts, a page for each. That's what I wanted. And I figured an AI coding agent should make rebuilding it almost free. So I pointed OpenCode at it and asked it to build me one. I started from a fresh `laravel new`, the same way I start any project. What I didn't expect was that I'd end up taking more out than I put in. The trade-off isn't free. Skipping a commercial platform means the server and the app are mine to keep alive, the same maintenance that made me walk away before. I'm OK with it. What's different now is that coming back doesn't mean starting over. Here's what happened. I didn't design a blog. I pointed my agent at one, and that's the whole method. What came back surprised me. It didn't just copy the layout. It came back as Herman's blog. Herman is the person behind the Bearblog site I used as a reference, and the agent pulled his real content: his name on the about page, his projects, his posts, the newest ten in full. A working prototype of someone else's site, running on my local machine. The agent copies what you point it at, and I'd pointed it at his. His posts turned out to be useful scaffolding. I dropped my own draft into the same shape. From there it was just steering. Replace his content with mine, drop his posts. Where the agent hadn't built something, it left a placeholder, like a `#` where the XML feed was supposed to go, and let me decide. I asked for the feed and it built it. I skipped the contact page and used a plain mailto link. I picked what to keep. It even interviewed me. To fill in my about and "now" pages, I had it ask me questions: my name, what I do, what I'm working on. Talking it out with something, even an agent, surfaced content I couldn't just write cold. Bearblog's approach is CSS embedded right in the layout, no external stylesheet. So my layout just doesn't load the Tailwind `app.css` Laravel ships with. Posts aren't in a database either. They're flat Blade files, just HTML, no CMS, no components. My production footprint ended up smaller than my starting point. I took more out than I put in. That's also why I think this one sticks: less moving parts means less to break and less to keep updated. No database to back up, no CMS to patch. The upkeep stays small because the blog itself is small. And that's the whole bet. The thing that killed every blog of mine wasn't the abandoning. It was the cost of starting fresh. A blog this small is cheap to rebuild. Deploying was the easy part, because my VPS was already set up. I run Eddy, a self-hosted server management panel, alongside other Laravel projects. Adding the blog meant creating a site, pointing it at the repo, and deploying. The setup is the only hard part of self-hosting, and I'd already done it. Start to live was maybe an hour. I don't remember exactly, and that's the point. It moved fast enough that I didn't have time to procrastinate before the post was up. If you want to do something like this, there's not much to it. Point your agent at a reference you like, let it copy, and you have something to work from. Take out what you don't need, add what you do. Publishing works the same way. To put up a new post, I point my agent at an existing one as a reference. It writes the new one in the same shape, adds the slug to my blog controller, commits, and pushes. Then I open Eddy and click deploy. That's it. I could set up a GitHub Action to auto-deploy on push, but I haven't yet. Manual first, automate later. Same way I ship the writing. The real thing that kills a blog isn't the upkeep. It's whether you keep showing up. So make it a place you actually want to come back to. What I know works is pointing an agent at something you like and letting it copy. I'm not sure what else they're good at, but copying works. And maybe the bigger thing. It's fine for stuff to end. I publish now. If this blog rots someday, I'll make another one in an afternoon. I don't need it to last. I just need to be able to make it again.

• 
  brick (@werdcel) reported 29 minutes ago

  @igama GitHub is old fashioned because it’s made for humans to look at and review code. It solves problems that are going away. In five years it will be laughable to use something like what GH is today. It’s just going to be a headless service agents use.

• 
  Nitesh (@NiteshTechAI) reported 29 minutes ago

  Uptime Robot and Better Uptime charge you monthly just to watch your own server. This one self-hosts for free and you keep the data. It's called Checkmate. • Status pages with four themes built in. • Slack, Discord, PagerDuty, and SMS alerts. • Self-hosts on a Raspberry Pi or your own server. • Stress-tested past 1000 monitors with no slowdown. • Uptime, Docker, ping, SSL, port, and game server checks. Eleven notification channels means it slots into whatever your team already uses instead of one more dashboard nobody opens. ⭐ 10,000+ stars on GitHub. AGPL-3.0 licensed. 🔗 GitHub link in the comments 👇

• 
  ZomboDB (@zombodb) reported 31 minutes ago

  Whenever GitHub is down I smoke meat.

• 
  Daniel (@MnFounder) reported 35 minutes ago

  @johniosifov I have 8 agents and I don't give the same permissions for all of them, in fact that was one of the reasons why I split into 8 instead of having multiple. My coder agent is the only one with github write permission, my designer is the only one with access to canva, and so on. This way if something breaks it is easier to pinpoint the problem and fix it.