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:

• 
  Aditya S2 (@AdityaPlusp) reported 3 minutes ago

  @readylayerone github link is not working? ig

• 
  NakCrypto (@Rpequ2322) reported 3 minutes ago

  Been skeptical of $NOCK's demand story for weeks. Fair is fair though: I scanned the GitHub again and the *** looks good. What changed this week: The AI-PoW branch went from a month stale to a full adversarial audit sprint on July 8. Not vague commits, real attack-surface work: MoE expert-routing bleed mitigations, noise-matrix pinning, matmul under-constraint checks, verdict tables reading "11 SAFE, 1 issue fixed." That's pre-launch hardening behavior. You don't audit like that unless you're about to ship something adversaries will attack for money. The loop speed is the wild part. Founder posts "working on MoE support" July 7. Community asks sharp questions about tile selection and expert routing in TG the same day. July 8, the repo shows audit commits against exactly those surfaces. Question to code in 24 hours. The PR went 484 → 543 commits in a week. Pearl merge-mining compatibility actively maintained ("matching Pearl, fork fix"). Even the block-size tweet has fakenet test commits behind it, not just vibes. Small team, basically one workhorse committer carrying master. That's a real risk. But the work is disciplined, adversarially audited, and matches the founder's public claims commit for commit. My skepticism was never about the engineering. It's that every commit hardens the supply side, the puzzle, and nothing yet touches paid demand. That question is still open. But the launch looks real and close, and the code quality is exactly what you'd want to see. Credit where due.

• 
  Yazi (@Yazi_27) reported 3 minutes ago

  @bil0090 Well im sure I tested with workflows too, but this issue has been widely reported on github claude code, I should maybe try and run it again.

• 
  Michael (@MasterMike88) reported 4 minutes ago

  @icrazeios Dear @github, fix this ****, thanks.

• 
  Vatsalpandya333 (@Vatsalpandya333) reported 4 minutes ago

  Production bugs are not just engineering problems. They are customer-retention events. A customer reports an issue. The team searches Slack, logs, Sentry, GitHub, and deploys. Hours later, the bug may be fixed. But the customer is still waiting. The real problem is not just the bug. It is everything that happens after. Customer report → investigation → root cause → safe fix → customer follow-up One context. One timeline. One workflow. That is what we are building at @TasksMind.

• 
  FilliDeFilla (@FilliDeFilla) reported 4 minutes ago

  One thing I like about open source software is being able to turn "I wish this worked differently" into a GitHub issue

• 
  Kevin Hurley (@kphur) reported 7 minutes ago

  There's been a round of misinformation about Spark going around, so for the sake of setting the record straight, I'll briefly clear up a few things. For one, unilateral exit has been around since the early days of Spark. Many developers and users have used it. This has been demonstrated many times both here on X and during the process of integration by developers. Unilateral exit also does not require the SOs to be online when a user wishes to exit. When a transaction is received, users can save the unilateral exit information and later use those pre-signed, valid L1 Bitcoin transactions at any time on Bitcoin. There are existing Github issues to expose unilateral exits in a more intuitive way in the SDKs, but unilateral exits themselves have been functional for a very long time. Unilateral exits do require CPFP - this is used to ensure that the expected value for an attacker is negative. The typical user would perform a cooperative exit, which does not require any on-chain funds and is an atomic swap of on-chain funds in exchange for Spark funds. Unilateral exits are generally reserved for a worst-case scenario and can be sponsored by an L1 fund provider if needed. Second, the confusion around "Sparkcore". At Lightspark, we use a monorepo for our server code. This one service is called Sparkcore - the naming of which preceded the creation of Spark. Lightspark runs an SSP within this service. Our Lightning infrastructure uses both LDK and LND - both of which we contribute code towards. Sparkcore itself is not open sourced - that would mean open sourcing our entire server-side stack for every product we have built. The Spark network code, however, has always been open source - and that's the openness that matters, because it's the code that actually enforces the rules of Spark. The SSP is an optional, replaceable convenience role. A recent post claimed that APIs used for other products are part of the SSP. We have many products, and we have never been shy about describing UMA, which allows regulated entities to exchange information to process transactions over Lightning. This is not a Spark product. The SSP does not hold your seed phrase (that should never leave your device), the SSP cannot freeze your funds, and the SSP isn't even a required role to use Spark - it is the interop layer between Lightning and Spark and helps do swaps for exact denominations of leaves. Running an SSP is something we have talked with many partners about. The client chooses which SSP they wish to interact with (if any) - we cannot control if a client talks to a new SSP. Finally, privacy. I've discussed this many times in the past, so won't belabor the point again. Spark allows for transactions to be hidden from external visibility. As I've spoken about at length both here and at various conferences, we care deeply about making sure that there is true privacy, and we aren't satisfied with anything short of that. It's an ongoing effort to continue to further the research in this area. I'll leave it with this. In the network our critics operate, the default payment path is one where the operator colluding with any prior owner can double-spend the current holder - their own docs say so. Receiving over Lightning means trusting that the operator deleted a key - their own docs say so. If you don't come online every 28 days, the operator can take your funds. In their founder's own words: "In theory it could steal it." The automatic re-issuance of expired funds promised in March 2025 still hasn't shipped. Their operator's liquidity costs scale with payment volume, which by their own admission "will translate into user fees." And there is exactly one operator - their own docs tell everyone else: "Do not attempt to run an Ark server in production (yet!)." Spark has three independent operators, exits that don't expire, and no flow where a single operator can take user funds. Users can judge for themselves. Our users and the developers building on top of Spark care about bringing Bitcoin to more people. They value the ease of use and simplicity of Spark. They care that we have 3 independent SOs. They care that we are pushing for more and better functionality. And they value that we spend all of our time thinking about how to make Spark better each and every day. Ok, now back to building because that's what we do at Spark.

• 
  Glenn 'devalias' Grant (@_devalias) reported 8 minutes ago

  @thsottiaux IMO it should be a default part of the repo's agent instructions / GitHub actions / similar that raised PR's should explicitly cross-link to related issues raised; ideally with 'closing keywords' / etc so that GitHub's awareness features can actually work as intended.

• 
  Gregor (@bygregorr) reported 14 minutes ago

  @github The fields aren't the bottleneck. Half my issues have no priority set even with labels and milestones already there.

• 
  Aaron Delasy (@aarondelasy) reported 15 minutes ago

  @zeeg somehow it seems agents are really bad at configs, they have to test and re-test everything multiple times to get it to work. I remember I had a problem in github actions, and for a single line change it took agent around an hour and 300k tokens

• 
  Kaspa Daily (@DailyKaspa) reported 18 minutes ago

  Two weeks since Toccata went live on Kaspa mainnet. I checked the actual developer numbers instead of the vibes. Here's what the data says: — New Kaspa repos on GitHub: 39 in July 1–14 alone, vs 58 in all of June. Fastest monthly pace this year (March was 52, April 78, May 70). — Covenant-specific repos running at roughly 2x the pre-fork rate. — Silverscript: 21 forks against 42 stars, a 1:2 ratio means people are cloning to build, not bookmarking. 15 PRs/issues in the last weeks, and external contributors are now landing code: a Groth16 verifier builtin, typed sig-check builtins, an RFC for cross-contract validation. One issue is literally titled "from building a mainnet contract." That's the signal you want, outsiders hitting real problems and reporting back. What actually shipped in 14 days: the first covenant explorer (kascov), a covenant-based KAS vault, a native L1 covenant token, a covenant pattern library, a wallet standard, a Swift SDK, a testnet raffle dApp, several other projects under development. Most interesting pattern: three independent projects converged on the same idea, covenants as spending guardrails for AI agents. An x402 payment protocol binding, two agent wallets where the AI can only spend inside covenant constraints. And the community just voted $25K toward an AI agent hackathon at Imperial College targeting 1,000+ devs. The agentic-payments thesis is forming bottom-up. Core isn't idle either: Silverscript pushed commits this week, template hash hardening, reproducible builds. That's pre-production housekeeping, not feature chasing. Meanwhile discussion has shifted from price to fundamentals: the $6M developer fund and covenant atomic swaps are the topics now. Caveats, because they matter: Silverscript is unaudited and still landing breaking changes. Devs report RPC friction on deployment, up to 11 retries in some cases. And absolute numbers are small: this is dozens of motivated builders, not thousands. No major outside team has announced a covenant product yet. But two weeks in, the shape is clear: infrastructure activated, tooling hardening, and builders showed up without being paid to. The Q3 question is whether that compounds.

• 
  Onions Gillespie (@maxcsmith) reported 18 minutes ago

  This isn't a pitch it's just what will be in its modular setup. Other engineers have no trouble compiling from the Tom A. *** Notes. Tom like Tom Hanks or Tom Cruise, but any Tom- not after me, Tom A. "Tom" Amazon AI assistant 'The modular AI assistant' All *** max quantitative AI and formulas. Ready for github. Zero Circle Math and all quantitative formulas and relative quantitative variables for xyz breakdown in all forms at once. Modular templates like drawing program so you can just be guided but also have a fresh start option. Pick a quantitative breakdown. Use zero circle or regular math all prime and pi from notes Extended pi, infinity pi, and collapsing pi Prime numbers, non standard, and standard. Program modulars with templates. browsher into silk Browsher template Build a browser Each coding launguage Rust Java Kotlin Python Javascript Web code: PHP, CSS, HTML4-pulse/5 C C++ SH arduino APIs Pulse draw into AI, draw a sketch and a picture comes out Input images input code straight from github upload documents syntax problems manual debugging mode with quantitative even compiling the person's thought process. Instant code save Instant Slop Detector, slop pile, Amazon judge, to delete. Can save. Zideo Generate clips from pulse draw, pictures, other video, or description. No copyritten files off Amazon. Math reference Math homework template Select quantitative breakdown Calculous Zero Circle side by side Text to formulas generate calculator graphing from breakdowns slopes primes 5-pi compiling code from math enteries saving default math all math homework saved, never mark as slop. Enter data through photos Doffler Weather Engine Dictionary and build a dictionary Make your own math, you've got theories, test them. All quantitative has been mapped. Quantitative award if found, there won't be one. Forstall like Philosophy to math Logic. Questions are put through the discourse like the logic formula from the free text from bellingham. Tom bias rating. Where tom has bias, it'll admit. Provides a theory behind the bias. "What's the bias meter?" Video Game Template. Build a game! Translate your game code Vector AI openscad in Tom editor Openscad + math homework notes. Ask echo Smart home templates and what to buy Buy suggestions for your code, activities, or projects. Pressure chem template Hortiquestions Assistant Gardening

• 
  Warizo (@Warizo_ofAfrica) reported 20 minutes ago

  @github Moving away from monolith models to a smart subagent delegation architecture is the real future of terminal agents. In the CLI, tool and search failures completely break engineering momentum, so cutting those errors by over 20% is a massive workflow win.

• 
  Nexis (@Nexisintel) reported 21 minutes ago

  A 13-year-old uploaded a free trading tool to GitHub. 60 days later, someone sent him $20,000. No course. No subscription. No paywall. He spent 14 days of his school vacation building a simple Polymarket terminal while his friends played video games. When it finally worked, he pushed the code to GitHub, went back to school, and forgot about it. Two months later he opened GitHub again. Hundreds of traders were using his project. One message stood out. A trader said the tool had helped him make over $200,000 in a single month. He asked for the kid's crypto wallet. Then $20,000 showed up. The internet rewards usefulness faster than marketing. The best distribution isn't always ads. Sometimes it's open-source code that solves a real problem. Build something people genuinely need. The right users will find it. And sometimes they'll pay you before you even ask.

• 
  chuplung (@choopyplug1) reported 29 minutes ago

  Claude found a security bug that humans missed for 27 years. Anthropic's full developer keynote: 6 moments from this keynote. the first 10 minutes alone are worth your time • 02:38 - Stripe had 50,000 lines of Scala to rewrite. estimated 10 engineering weeks. done in 4 days with Claude • 04:50 - Mythos found a 27-year-old vulnerability in OpenBSD that survived every human reviewer for 3 decades • 08:11 - SpaceX compute partnership announced. rate limits doubled across all plans • 35:36 - routines: set it up once, Claude kicks off work from GitHub issues, webhooks, or schedules while you sleep • 37:24 - MercadoLibre: 23,000 engineers on Claude Code. 500,000 PRs reviewed. targeting 90% autonomous coding by Q3 • 42:06 - Boris Cherny: "I'm not the one doing the prompting anymore. I'm the one creating a routine that does the prompting" save this before it gets buried ↓