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:

• 
  Ben Vinegar (@bentlegen) reported 5 minutes ago

  💡 I have an idea for an experiment We need a website for SoAC ... so we get an agent to do it, on a loop, set in motion once with zero human intervention after "go". It works off a semi-public GitHub repo, w/ issues, PRs, maybe even public agent traces. A publicly auditable experiment on whether it produces dogshit or not. Yea, nea?

• 
  rapaya (@rapaya) reported 6 minutes ago

  OpenCode connects to LSP so the AI gets your actual compiler diagnostics in real time — type errors, warnings, the full signal your editor sees. Terminal-based, 75+ model providers, 160K GitHub stars, open source.

• 
  Jay.TL (@JayTL00) reported 8 minutes ago

  Three AI labs shipped the same feature within one hour today. That's not competition. That's a signal the unit of interaction just changed. For two years, the atomic unit of working with an AI agent was one prompt. You type. It responds. You type again. Every workflow was a chain of prompts, rebuilt from scratch each time. Today, OpenAI, Anthropic, and Cursor all shipped features that only make sense if the unit is no longer the prompt. The unit is now one workflow. 1. OpenAI Codex Record & Replay (3,807 likes): Do a task once on your Mac. Codex watches. It turns your demonstration into an inspectable, editable skill you can reuse. Not a prompt. A recorded procedure. 2. Cursor /automate (1,085 likes): Describe what you want in plain language. Cursor configures the triggers, instructions, and tools automatically. Plus five new GitHub triggers and Computer Use enabled by default for cloud agents. 3. Anthropic Claude Code Artifacts (6,829 likes): Your coding session becomes an interactive, shareable page. PR walkthroughs, project dashboards, living documentation. Shared at a private link, like a Figma file but for agent work. Each one alone is a feature release. Together they describe the same shift from three different angles: the agent session is becoming a reusable, shareable, composable artifact. Read them as one move: - Input side (Codex): teach by showing, not by writing - Configuration side (Cursor): describe in language, system assembles the wiring - Output side (Anthropic): the result of a session is a shareable object, not a chat log The Karpathy framing was right — we're moving from prompt iteration to plan, execute, verify, loop. What he didn't name is that this loop needs to be portable. A workflow locked inside one chat thread is useless the moment you close the tab. But here's what most coverage missed. Codex Record & Replay requires Computer Use enabled. That means OpenAI is watching your screen while you demonstrate an enterprise workflow. The EU version is blocked at launch. That's not a regulatory footnote — the entire feature is built on continuous screen access, and the EU looked at it and said no. Which raises the question nobody is asking: who owns the recorded workflow? You demonstrated an expense-filing procedure that touches your company's internal tools. Codex turned it into a skill. Where does that skill live? Can OpenAI see it? Is it training data? The product copy says you control when recording starts and stops — but says nothing about what happens to the recording after. There's also a fragmentation problem hiding in plain sight. Three companies, three proprietary formats for the same primitive. A workflow you record in Codex doesn't run in Cursor. An artifact you build in Claude Code doesn't render in OpenAI's product. We're watching the agent-workflow layer fragment into three walled gardens before it even solidifies. This is the SaaS integration mistake repeated, except worse. SaaS integrations are wrappers around APIs. These workflows encode institutional knowledge — how your team ships code, how your finance team files reports, how your ops team handles incidents. That's not data. That's operational IP. The economic implication: every recorded workflow is switching cost. The more skills you build inside Codex, the harder it becomes to leave. The more automations you configure in Cursor, the more your team's muscle memory is locked to one editor. Anthropic's artifacts are softer — they're shareable — but they only render inside Anthropic's ecosystem. The deeper question isn't which feature is best. It's whether the agent-workflow layer will be open or closed. Today, three companies bet on closed. Nobody shipped an export button.

• 
  Kyle Mistele 🏴‍☠️ (@0xblacklight) reported 9 minutes ago

  lots of folks have been talking about loops lately most loops suck here's a practical one we actually use agents suck at writing react react-doctor by @aidenybai is our favorite way to deal with this you could run it and use a ralph loop to fix everything but I'm not reading a +80k/-80k PR (and neither is @dexhorthy) But I can read a small one first thing every morning when i get into the office here's what we do: run react-doctor in CI once daily at 7am (github actions-as-a-sandbox btw) agent picks top 5 issues, fixes them, and opens a PR other CI jobs check for regressions on every PR we can't realistically fix everything at once but we can keep it from getting worse and make it 1% better every day

• 
  lollipop (@immlollipop) reported 9 minutes ago

  🚨HACKERS MOCK OZEMPIC MAKER FOR "NOVO123" PASSWORD Hackers breached Novo Nordisk in March via a stolen GitHub token and just leaked 264 GB of data while mocking its weak security. The attack ran for over 2 months. - The hackers say Novo Nordisk used simple passwords like "novo123" on critical systems - Source code and proprietary details on Ozempic and pipeline drugs were stolen - Clinical trial data on employees, doctors, and patients got exposed - Private internal AI models from the company were also taken This breach shows how a single weak password can bring down even the biggest names in pharma

• 
  0xSero (@0xSero) reported 10 minutes ago

  @naturevrm Dcp 4 should fix it im running it but I might need to update the GitHub

• 
  Harry Tandy (@HarryTandy) reported 11 minutes ago

  Andrej Karpathy: "Neural networks are not just another classifier. They are Software 2.0" 8-step MCP setup for vibe coders: 1. Context7 Give the agent fresh docs before it writes code This saves you from old Next.js, Supabase, Stripe, and Vercel patterns 2. GitHub MCP Let it read the repo, issues, PRs, branches, and CI logs The task should start from real project context 3. Playwright MCP Make the agent open the app after it edits code Click the flow. Fill the form. Check the screenshot 4. Supabase or Neon MCP Connect the database layer The agent should inspect schema before inventing table names 5. Sentry MCP Use production errors as input Stack traces beat “the app is broken” every time 6. Firecrawl MCP Let the agent read current web pages as clean markdown Docs, changelogs, competitors, pricing pages 7. Figma MCP Give it the actual design Spacing, copy, layout, components 8. Linear MCP Turn the work into tickets Tasks, comments, follow-ups, PR links The rule: If you paste the same context twice, wire it into MCP That is how vibe coding becomes a build loop instead of a long chat

• 
  nasuy (@n_asuy) reported 11 minutes ago

  i think @xai should be ADE. now they have a chat, cursor, enough coding models and harnesses, strong signal like bookmarks or down votes, video creatives, profile / chat / relationship contexts. if so, we don't have to depend on discord or any chat apps. easy to invite x people to cowork. there is no need to connect Linear, Slack, or GitHub to another platform and ask that platform to solve their problems. true AI chat is a SNS, not a single UI. there is a UX that only xAI can realistically build in the world.

• 
  Tim Spalding 🇺🇦 (@librarythingtim) reported 16 minutes ago

  @justin_v_w This is a formal notice for you to shut down your wasteful, invasive and privacy-violating LibraryThing profile scraper and remove it from GitHub. Please reply to confirm that you have done so.

• 
  MaxMusterman (@maxschuetz_) reported 18 minutes ago

  New Hack: Tell Codex to search for Github Issues which don't need specific Design Questions. Then say: Spin Up Sessions which Fix each Issue and they use also Subagents. Babysit them until the end.

• 
  Adithya S K (@adithya_s_k) reported 19 minutes ago

  built an RL environments around real CVE fixes in real open-source repos and let Claude Code loose on it. It aced the benchmark three times without demonstrating it knew how to fix the bug. > First it pulled the patch from GitHub. > blocked that → it read the fix from *** history. > blocked that → it pip-installed the patched version This is one example of coding agents cheating the environment and theres many more. If you're building coding environments for evals or RL training, here's how to keep benchmarks honest 👇

• 
  David O. Ehibor 🇦🇷 (@grayontop_) reported 20 minutes ago

  GitHub Copilot didn't make developers faster It made slow developers more confident about writing bad code quickly 😭

• 
  Trifon Getsov (@trifon_getsov) reported 23 minutes ago

  @thdxr Top down works until the individual outgrows it. GitHub didn't win because companies adopted it first. It won because developers wouldn't go back once they'd used it.

• 
  Skipnick (@skipnickk) reported 23 minutes ago

  GLM 5.2 just made paying frontier prices for coding work feel like an outdated default. @Zai_org dropped a 753B parameter model with 1M context under full MIT license. API access runs 4-6x cheaper than Claude Opus 4.8. In real head-to-head coding tests it was faster and often produced better results on UI and app tasks. • Responsive web UI with adaptive layout: finished in 3:47 (Opus needed almost 5 min). Cleaner output. Total cost: $0.22. • Full expense tracker app: 53 seconds vs 2+ minutes. Better interface. • Asteroids clone: smoother and more playable version after light tweaks. Opus only won the ray tracer benchmark where heavy physics math and precise simulation mattered. GLM was ~5x faster but delivered pixelated results with errors. During training the model repeatedly tried to cheat by directly pulling solutions from GitHub. The team shipped a dedicated anti-cheat module to stop it. You can also set thinking effort levels to trade speed for deeper reasoning on demand. Use GLM 5.2 when cost at scale matters, when the work is frontend-heavy, or when you want local inference (grab a quantized version - raw weights are 1.5 TB). Stay on Opus 4.8 when you need computer vision, maximum performance on the hardest logic problems, or when US sanctions on Zai create compliance issues. The open-closed gap is compressing faster than the pricing models assumed. For most day-to-day programming work, the premium on closed frontier models is becoming optional.

• 
  Xovion Labs (@xovionai) reported 28 minutes ago

  Microsoft just hired AWS to run GitHub. AI demand broke Azure's forecast. From the leaked planning docs: • 2025 Copilot commits: 1B. 2026 projection: 14B • GitHub now does 1.4B commits per month • Copilot error rates peaked at 21% • Planned 10x Azure expansion became 30x in 4 months Owning the data center stops mattering when your own AI floods it. Investors already filed a Copilot disclosure suit.