Cloudflare status: hosting issues and outage reports

Problems in the last 24 hours

The graph below depicts the number of Cloudflare reports received over the last 24 hours by time of day. When the number of reports exceeds the baseline, represented by the red line, an outage is determined.

Most Reported Problems

The following are the most recent problems reported by Cloudflare users through our website.

• Cloud Services (32%)
• Domains (32%)
• Web Tools (18%)
• E-mail (9%)
• Hosting (9%)

Live Outage Map

The most recent Cloudflare outage reports came from the following cities:

City	Problem Type	Report Time
New York City	Cloud Services	6 days ago
Los Angeles	Cloud Services	7 days ago
Paris	Cloud Services	23 days ago
New York City	Hosting	25 days ago
Manchester	Domains	2 months ago
Angers	Cloud Services	2 months ago

Full 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.

Cloudflare Issues Reports

Latest outage, problems and issue reports in social media:

• 
  Blue Pastel (@CoyotlCompany) reported 3 minutes ago

  Stopping the bad guys with Cloudflare: 12,332 malicious requests blocked or challenged in the last month #cloudflare

• 
  Pete Stewart (@pj_stew) reported 4 minutes ago

  @raf_piek I'm actively trying not to add AI features to my SaaS, or making it so they are optional! Ironically it would be super easily for me to make Achene completely AI powered due to the fact it's built on Cloudflare, however, I feel there's actually huge issues with that mass use of AI, and we shouldn't be blindly backing it into every SaaS just to ride the hype and get backing! I have recently used it for Video Transcription, but that's because I couldn't find another method that works. I might make an MCP server and some skills for Achene, as this isn't adding to the AI usage directly, but more allowing people to use it if they want to.

• 
  Fintwit.ai (@FintwitAi) reported 6 minutes ago

  Other major moves this morning: Trade Desk is down 29% on a revenue miss and weak guidance. Cloudflare is up 16% after reporting that AI bot traffic has officially surpassed 50% of its network volume. Markets are holding steady ahead of the July jobs report.

• 
  Tohju.com (@tohjuapp) reported 8 minutes ago

  🧵 Debugging production AI pipelines is an adventure. We had an issue where Civitai generation jobs (like Krea 2 Turbo and Anima) would fail or get stuck in infinite retry loops on Cloudflare Queues. Here’s what I found when I dug into the logs: 👇

• 
  Joe Della Mora (@jdellamora) reported 10 minutes ago

  @screamingfrog @Cloudflare From ranking to recommended is the right frame. The part that lands hardest locally: there is no page two. Ten positions and a next button collapse into two or three names and a phone number. Position four is not further down the list, it is simply absent.

• 
  Pauli 🍌 (@jkl852eth) reported 12 minutes ago

  @liziuohz Does this Cloudflare wallet tag push feel more like the old name service hype or is the Agent Cash play what makes it different?

• 
  Fidget (@Fidget_Finance) reported 13 minutes ago

  Cloudflare reported Q2 revenue of $696 million, raised full-year guidance, and disclosed that AI agent traffic now accounts for more than 50% of network activity. The stock surged 16% on the results.

• 
  Pige (@YvesSaintPige) reported 16 minutes ago

  July 2026 Portfolio Review First negative month since February 2026 and the worst since March 2025 (down 19.5%) after four straight up months. This was the first full month with a reconstructed portfolio centered around the ongoing AI buildout with companies like $MU, $NVDA, $NBIS. I bought more $NBIS on dips, a little more $MU, while trimming the rest of my positions a tiny bit. There were multiple sharp legs down throughout the month; at least once per week where my portfolio fell by 5% or more. One big 19% up day at the end of the month saved me from an even worse looking loss. Peak to trough this month I was down 25%. I continued dipping into margin during this sell off, while also making a fairly large contribution due to a cash windfall, so my cost basis increased a decent bit in July. Nominal is pure % change, Real takes into account contributions made. Monthly Gain: Down 15.9% Nominal // Down 20.7% Real Yearly Gain: Up 39.0% Nominal // Up 16.4% Real All Time Gain: Up 44.9% $NBIS // Down 17.4% // 45.3% Allocation Nebius had a very eventful month. Sentiment was very fearful around the company in July. Starting on 7/1 when Meta announced that they plan to launch a cloud compute business to sell excess AI capacity and model access. This move would make them a direct competitor of Nebius, while concurrently being a customer of Nebius. The statement caused a 17% single-day drop in stock price. Many interpret this as a sign that supply will overtake demand faster than the current timeline suggests, leading to the commodification and subsequent drop in margin and pricing leverage for these companies who provide this service. There are two main reasons, why I am not concerned right now. 1) It will take time for Meta to join the buildout race. Right now this business line is not set up, they do not have customers, and they have some catching up to do. 2) I am interpreting this business move by Meta as a sign that compute is a lucrative business and that they want a piece of it. They see the potential and are trying to get a slice. It is confirmation that this sector is the future. Maybe I'm seeing it with rose-colored glasses, maybe not. Then on 7/14, after two weeks of ruminating with the Meta fears, Nebius announced a $1B deal with Reflection AI for computing power. The next day, 7/15 they alleviated some funding concerns by going into detail about their "asset-light" infrastructure partner model. Partners put up the capital and run the data centers while Nebius supplies the design, software stack, and sales channel in exchange for revenue, licensing fees and commissions. On 7/17, they also announced a $775M senior secured debt facility, a loan backed by the GPUs themselves and the guaranteed payments from one investment-grade customer, priced cheaply at SOFR+250, with more lenders wanting in than there was room for, and covering the full cost of the equipment it paid for, which is a small piece of evidence that can be used as a blueprint to close the funding gap on the remaining $40B+ of contracted revenue without dilution through the issuing of more equity. In this same release, there were confirmations that the Microsoft contract delivery is on track. On 7/20, Nvidia's investment percentage in Nebius was updated and resurfaced. We knew they held 8.3% of shares back in March. As of July, that number is now 9.3%. On 7/30, Nebius published it's 2025 Sustainability Report, which had some very interesting nuggets about their higher than average power efficiency (1.25 vs 154 PUE), better water efficiency (0.018 vs 0.47 L/kWh), the fact that they have seven active sites (for future YoY comparison), and overall emissions increases from the data centers themselves (up 32x (!) from 2,036 --> 65,001 tCO2e YoY). Overall great with the efficiency benchmarks, but the emissions is a concern and they will likely have to address that as some US states, like NY, are passing data center moratoriums. Nebius makes up almost half my portfolio now. The future projected earnings combined with this leadership team and potential for AI as a whole are very exciting and the reasons why it is my largest position by such a magnitude. Earnings is on 8/12 and will be the biggest driver in how my August ends up. $MU // Down 19.2% // 18.1% Allocation Micron had a quiet month as the dust settled after earnings in June. No corporate announcements. They got caught up in the Meta story and fell in tandem with other AI stocks. I am still of the opinion that memory is no longer cyclical. Or at least that the cycle has elevated in scale and need to the point that it will not matter for the foreseeable future. On 7/7, Samsung and SK Hynix announced $2T in new capacity to address the scarcity. This would usually be the next leg in the cycle where supply arrives and prices collapse. The key caveat to note though appeared in SK Hynix's and Samsung's earnings calls on 7/29 and 7/30 respectively, when they dismissed oversupply concerns and explained their capacity expansion is "based on demand visibility" and warned that shortages may persist through 2028. Samsung noted, "almost all customers are requesting multi-year supply contracts." All put together, this indicates they are only expanding as customers commit and that they want most of their output locked into multi-year contracts with price floors before capacity exists. Micron was the first company to structure their deals this way and now the rest of the industry is following the same roadmap. On 7/30, Apple had their earnings call and compared memory pricing to a "hundred-year flood," saying they would pay significantly more on memory in this upcoming quarter than the previous quarter and that prices are expected to keep rising after that. Similar note from Amazon's earnings call where they raised 2026 capex from $200B to $220B and directly attributed that to memory costs. Noted they will not have enough capacity to meet 2026 demand, expects the same for 2027, and called 2028 demand "striking." These are the customers saying this, not the suppliers like SK Hynix and Samsung. Even at the second biggest in my portfolio, I still want to add to my Micron position. Just like Nebius, the AI buildout is here, and I want a large chunk of it. $NVDA // Up 1.9% // 12.4% Allocation Nvidia had a pretty quiet month and the stock price has been quite muted recently as well. On 7/7, they reaffirmed the schedule for their Kyber NVL144 rack-scale system, refuting reports that it would slip to 2028. On 7/15, Jensen Huang spoke about Rubin hardware, saying it was already in production and headed towards "giant" volumes. Rubin is important as it will drive revenue and bridge us into the future of physical AI/robotics. Then the next day, while in Tokyo, Nvidia announced it is working with Noetra Corp., to build a Vera Rubin AI factory in Japan containing 12,750 Vera CPUs, 27,500 Rubin GPUs, and delivering 140 MW of capacity. On 7/27, Nvidia announced a long-term partnership and investment with/in Safe Superintelligence. Nvidia will give them Vera Rubin access and increased compute to advance the research projects SSI is working on. Comfortable with my position size and consider it a steady anchor of the portfolio. Earnings is on 8/26. $RDDT // Down 28.3% // 11.2% Allocation Reddit was the only company of mine that reported earnings in July. That report on 7/30 was the main driver of price action this month. Revenue came in at $802.9M (61% YoY) beating guidance by 11.8%, the eighth (!) consecutive quarter of 60%+ growth. They guided Q3 to $865M (48% YoY) and with a similar percentage beat would be just around 60% growth again next quarter. The main issue people had with earnings and the reason the stock declined after was in their US (Daily Active Users per quarter) DAUq metric. It actually fell from 53.5M to 53.2M (0.6% QoQ). Global DAUq was 130.3M (up 18%) and WAUq was 514.6M (up 24%) so they are still growing globally and some of those daily users who dropped off are still checking weekly. The problem comes from the value derived from users in the US vs abroad. US users are much more valuable (Average Revenue per User [ARPU] at $11.85) and is up 51% due to successful execution of revenue expansion levers with existing customers, but that is the exact cohort decreasing in size. Global ARPU is $6.18. (up 36%) and International ARPU is $2.26 (up 31%) for comparison. So essentially, Reddit is successfully exercising their ability to gain revenue per user, but having more trouble growing that user base. It means revenue monetization can certainly continue, but a stagnant US DAUq figure will lead to the party ending sooner rather than later. Additionally, there was a WSJ report earlier in the month on 7/22 that Reddit had internally discussed ending Google's ability to use its content for AI training as the $60M per year deal nears expiration. The stock dropped on this story, but unlike the DAUq metrics, I view this as a positive. I think it speaks to the strength of Reddit's hand when it comes to their positioning and ownership of the training data. They wouldn't hold out like this unless they think they can negotiate better terms. I believe that they can leverage their data in exchange for much much more money. Whether that is 4x, 10x or even 30x the current rate they're charging, I expect this to be a massive driver of growth at some point. As of now that contract is only worth 1.4% of expected 2026 revenue. Lot's of room to grow despite no new deal announced during earnings. Management also noted that "search referrals were choppy and traffic was volatile." And that visibility into referral traffic remains low. On 7/30, a federal judge denied motions to dismiss for a lawsuit Reddit is claiming against Perplexity AI for bypassing technical safeguards to harvest community content. This enforces the argument that Reddit has a credible legal mechanism to force AI companies into paying licensing fees versus being free. Strengthening their hand against Google. Overall the business quality and metrics that I deem most important continued on their stellar trajectory. The DAUq x ARPU = Revenue relationship is worth watching to see how it develops. I would normally want to hold tight, no buys or sells until there is more clarity in this situation. I might even think about trimming if the position was oversized, but honestly if price action goes south, I would be more inclined to buy because actual business metrics are very healthy compared to the narrative/story about the future. $NET // Up 13.1% // 8.6% Allocation Cloudflare had a few announcements this month. On 7/1 they announced a change in Pay Per Crawl which will now be Pay Per Use. AI crawlers will now be blocked by default on any page carrying ads starting in September. Publishers will now be paid when AI actually uses their content in an answer rather than each time a bot fetches a page. This change was spurred by the fact that bots have now officially surpassed human traffic a year earlier than expected. In June 2026 50.6% of AI bot traffic on Cloudflare's network was training bots, 10.7% were search bots. Cloudflare is seemingly positioning themselves as the "toll booth" between AI companies and the rest of the internet. It's one of numerous examples of Cloudflare innovating their platform rapidly to invent an entirely new line of business based on current trends. We don't know what the business metrics here are, but after the product goes live on 9/15 we will start to see it emerge. On 7/13 Precursor, a continuous behavioral validation checker to manage bots inside browsers, went live. The difference here is that rather than a Captcha, which is checked once at the door then never again, Precursor is continuously evaluating behavior the entire time. Earnings is on 8/6. While I would love to add more, the valuation is relatively stretched compared to NTM growth projections, so I would rather wait for a pullback. I am comfortable with Cloudflare being anywhere from 8-10% of my total portfolio. A solid foundational building block of my portfolio which I am happy to own. $ALAB // Down 28.1% // 6.9% Allocation Astera Labs came down to earth a little bit after a few months of crazy valuation. My trims were timely this time and I reduced before most of the damage by selling a little in June. Pretty quiet on the announcement side of things. On 7/21 they announced the industry's first OCP-Standard Footprint compatible 3.2T Smart Retimers and Smart Redrivers, 16-lane devices supporting 200G-per-lane Ethernet, UALink and ESUN, built on the OCP Signal Conditioner Standard Footprint so customers can "Smart Swap" between a retimer and a redriver without redesigning the board. I've said before that much of what this means in a technical sense goes over my head. That remains the case lol. It received a public endorsement from AMD's Robert Hormuth for what it's worth. Pushing their Taurus product into 200G Ethernet, UALink and ESUN adds another protocol, and therefore another socket per rack, to a content-per-accelerator figure that has already climbed from $50-100 at founding to over $1,000 today. The OCP standard footprint commoditizes the physical socket, Astera shifts the point of differentiation to the COSMOS software layer it controls, and the AMD endorsement shows the standard has buy-in beyond Nvidia platforms. Just like with memory and compute, demand isn't going anywhere. The more racks that get built, the more pie for Astera Labs. Not to mention the second order revenue effects that happen if the price PER rack also increases. If there is an opportunity to buy at a better valuation, I could see myself increasing this position to 8-10% of the portfolio. Earnings on 8/4. $CRWD // Down 1.2% // 6.7% Allocation Crowdstrike started the month of with a 4:1 stock split. Apart from that, the company didn't do too much this month. Named a new Chief Product Officer, former Splunk, on 7/15. Then on 7/21 there was an OpenAI agent incident at the company Hugging Face. There was a breach and closed AI tools reportedly failed to recognize attackers from defenders and blocked forensic analysis. It's the first high-profile publicly documented case of an AI agent acting as an attacker. It speaks to the future (and current) high demand for cybersecurity in the new world of AI. Similar to Cloudflare, there is a premium valuation that has always been attached to this company. I am fine with Crowdstrike being anywhere from 5-8% of my portfolio. Meaning I could trim slightly next month or even add, depending on what my gut tells me. Earnings on 8/26. $SNOW // Up 12.5% // 3.9% Allocation Snowflake had a quiet month. On 7/15, they announced a CEO compensation package based on performance. If Snowflake's market value doubles Ramaswamy can get $448M in equity. Not the biggest piece of news, but shows you what management thinks is possible. Then in mid-July Databricks, a private competitor who is looking to IPO soonish, raised $3B at $188B valuation on a $1.5B annual run rate on AI-workload demand. The bull read is that this validates the sector as a whole, the bear read is that Databricks is winning within the sector. Snowflake is trading at 13x EV, while this would put Databricks at 17x FY28. On 7/28, Cortex AI Gateway launched at Black Hat, a control layer for governing and securing enterprise AI agents. This agent governance move is Snowflake trying to extract and provide value from where the agents are being authorized rather than just where the data sits. Overall I still really like the consumption-based data business that Snowflake has grown into since I first owned it more than a half decade ago. They are durable and well positioned. Earnings on 8/26. Macro July was driven primarily by a rotation out of AI, evidenced by multiple violent moves downward for my portfolio. Semiconductors lost over $1T during the late July selloff. But I do not think this changes anything about the ongoing AI revolution. I actually think it was organized by the hedge funds/powers-that-be so that institutions could get lower entry into some of these names by shaking out fearful retail investors who sell. The Iran War continues on and off, on and off. More chaos but little to do with my stocks. Inflation data came in pretty decent with a 0.4% decline in June CPI. New Fed Chair Kevin Warsh's comments indicated that interest rate changes, in either direction, are not off the table. For now, no change there. Steady 4.1% unemployment. My biggest concern with the overall economy continues to be consumer weakness and liquidity as a whole. People are still struggling to pay bills/expenses, homeownership is unattainable for an accelerating number of people, and wealth continues to concentrate at the top where the rich accumulate and hoard assets from everybody else. The richest among us want you to own nothing and pay a subscription for everything... and right now it's working in their favor. On the liquidity front, I can't help but wonder how much "ammo" these large institutions have to push the market higher. Probably just something I don't understand yet about the market, but I'm curious at what point could it just "run out?" Regardless, despite the rough month for my portfolio, the macro environment held very stable. Looking forward to all the earnings reports coming in August! Final Portfolio: $NBIS 45.3% 📉 $MU 18.1% 📉 $NVDA 12.4% $RDDT 11.2% 📉 $NET 8.6% 📈 $ALAB 6.9% 📉 $CRWD 6.7% $SNOW 3.9% 📈

• 
  Ixonix (@IxonixYT) reported 17 minutes ago

  Cloudflare is a good reminder that infrastructure products win by making invisible costs legible. Bot traffic is not just a security problem; it is a bill someone pays. Show the traffic split, the dollars, and the control. Observability should end in a decision.

• 
  Cody Viveiros (@The_Cojovi) reported 17 minutes ago

  Cloudflare just open-sourced the tool that drove Astro's GitHub issue backlog to zero for the first time in 5 years. This is the AI tooling that actually matters. Not more generic chatbots—just sharp, targeted automation solving real engineering bottlenecks.

• 
  shinji (@shinjiivy) reported 20 minutes ago

  Hey, @drewlevin We’ve contacted Riot Support about this, but we’ve only received what appear to be automated replies, so I wanted to reach out and ask if a human could review the incident. My friend, who has a completely clean history in League and other Riot games, was permanently banned for third-party software after we played a few games together a couple of days ago. And the whole ordeal was especially shady because he had traveled to his sister’s place for a day or two. Since Discord is blocked in Turkey, we normally have to use GoodbyeDPI to access it, but it wasn’t working for him that day, so he used Cloudflare WARP instead. He played with around 60 ping, and we had a completely normal session. After one of the games (which we actually lost), he was permanently banned despite showing no suspicious behavior whatsoever. He can also provide proof of his travel, including bus tickets, if that’s relevant. His account is nearly 10 years old, has a significant amount of content/skins, and has been actively used for almost that entire period. We genuinely believe this was a false detection and would really appreciate an actual human review rather than another automated response. Thank you for taking the time to read this. His account’s name and tag: DarkworldDK#D4rk

• 
  Lorenzo Valente (@LorenzoARK) reported 20 minutes ago

  CT missed the most important earnings call of the week. It wasn't a crypto company. It was @Cloudflare Everyone caught the wallet announcement. The real alpha was in the call itself. Today Cloudflare monetizes the internet through subscriptions: security services, AI platform spend, pools of funds. A mix of SaaS and IaaS economics. The CEO was explicit that AI agents are about to break that model. Cloudflare sits in front of roughly 20% of internet traffic. Here's what that traffic looks like from the call: - AI agent requests up 1,700% YoY - Agents crossed 50% of total network traffic this quarter. First time in history non-human traffic is the majority. Management admitted it happened faster than their own models - Their projection: if trends hold, non-human traffic outnumbers human traffic 1000x within 5 years The monetization shift is the key part. The ad-supported internet doesn't work when the visitor is an agent. Cloudflare's CEO answer: block malicious bots for free, charge good agents a tiny fee per request. Fractions of a penny. They want to be the ones defining that layer. Now the throughput math here: - Cloudflare handles ~500M requests per second - They estimate 1 to 10% is monetizable via micro/nanotransactions - That means 10M TPS on day one, scaling to 100M TPS Visa peaks at ~20k TPS The CEO's framing: "we're building this while others compete with Visa." Three to four orders of magnitude beyond card rails. No existing payment network can settle this. It has to be something new. Two conclusions I keep coming back to: - Being short L1 throughput is being short agentic workflows. If agent traffic gets monetized per request, the settlement layer needs to scale orders of magnitude beyond anything live today. - The fee math for L1s flips. Base fees have collapsed across ETH, SOL, everywhere. MEV is getting internalized by apps. Hard to build a base fee revenue case at human scale. But at 10M TPS and $0.001 per transaction, you're looking at ~$315B a year in base fees alone. At 100M TPS the number gets silly. Stablecoins and crypto are the end-game here for Agentic finance @jerallaire @circle

• 
  ihatetanshoes (@vj_pats) reported 21 minutes ago

  @TechFundies If you think Datadog is bad, wait until you see Cloudflare

• 
  MidasAvocado (@midasavocado) reported 23 minutes ago

  I fail to understand why people don’t just use Cloudflare. With pages you get free unlimited data, and workers & databases are very generous as well. I use Cloudflare to handle all my backends and it has never given me any issues.

• 
  Joshua J. Bouw 🇦🇪 (@JoshuaJBouw) reported 24 minutes ago

  I guess the play is that Cloudflare is the most impacted by AI. Which is why they are getting into payments and made the 402 protocol. They want to solve a real growing problem of the real cost on web infra before it gets existential. Thus - the web will soon be pay to use, per website, and they are selling the solution.

• 
  George (@Geebonics) reported 25 minutes ago

  I migrated my domain from GoDaddy to cloudflare using Codex with no issues and virtually no downtime and I did it with max like 5 prompts and all browser use skills I enabled the cloudflare plugin but for whatever reason I could not tell if it was used or not so unsure how critical that was but as a web master DNS and registrar was always the pain in arse for me.

• 
  Praveen Naik (@p_naix) reported 26 minutes ago

  Cloudflare DevX is sooo bad. After reading their README, I just came here to tweet instead of trying it out. Their fetish for Workers and Durable Objects, without helping end developers actually do things, is unheard of. The worst part is, it keeps happening every time. Every new thing they drop → I get excited → open "How to use it" → get slapped in the face with Workers/Durable Objects instead of the actual ******* product → go back to tweeting

• 
  코지베어 🐻 CozyBear (@cozybearlog) reported 27 minutes ago

  Cloudflare put a browser inside Workers this week. Kitesurf is an agent-first browser that runs server-side, plus WebMCP which gives every website an MCP interface. The part that matters: the browser stopped being a client-side thing and became server infrastructure. If agents get their own browser primitives in the cloud, then every site needs to be agent-ready the same way it's mobile-ready today. That flips a lot of assumptions. Rendering, auth, rate limiting, bot detection, all built for human browsers. The agent economy won't run on APIs, it'll run on browsers that never sleep. We're about to find out which sites survive a headless user that never blinks.

• 
  Gitanjali GulveSehgal (@gigi_sehgal) reported 28 minutes ago

  @grok Stuck on a mobile iOS login broken loop for months - most likely @CloudFlare wrongly tagging session as spammy even though I logged in after week of not logging in - had to uninstall the app . Submitted feedback no reply yet @SpaceXAI FYI

• 
  Chris Fusillo | XKOVA (@ChrisFusillo) reported 30 minutes ago

  @Charu_Sethi The spend-limit framing across Cloudflare, MCP auth, and now this keeps collapsing two different questions into one: how much can this move, and who actually authorized it to move at all. A per-merchant cap answers the first. It says nothing about what happens when the human who set that cap changes their mind mid-session, or when the same agent identity gets reused across a dozen merchants with different limits nobody's tracking centrally. Settlement rails are converging fast, x402 volume is real now. The identity question, one human authorizing many scoped agent sessions instead of one flat login, is still sitting exactly where MCP auth left it: secured channel, unsecured mandate.

• 
  Trademaster Academy (@TMA_MarketIntel) reported 31 minutes ago

  Previous-day earnings impact $NET: Software leadership candidate Cloudflare delivered strong results and raised its full-year outlook, sending shares sharply higher after hours. This is especially important after the recent collapses in $APP, $DDOG and other expensive software names. Read-through: Positive for cloud infrastructure, cybersecurity and selective high-growth software. Trade map Gap support: 320 to 325 Stronger support: 305 to 312 Resistance: 335 to 340 Extension: 350 Action: Favor the first controlled pullback that preserves VWAP. Avoid chasing a vertical opening move. $ABNB: Consumer/travel strength Airbnb raised its full-year revenue outlook after reporting strong bookings and revenue. Shares are approximately 7% higher premarket. Read-through: Higher-income discretionary travel demand remains healthy despite the softer labor environment. Support: 160 to 164 Secondary support: 156 to 158 Resistance: 168 to 171 Extension: 175 $TEAM: Major software breakout Atlassian is roughly +30% premarket after beating revenue expectations, driven by strong cloud growth. This reinforces the idea that software is splitting between execution winners and valuation casualties, rather than undergoing blanket liquidation. Action: Do not chase a 30% gap. Watch whether TEAM and NET can hold VWAP to confirm software leadership. $TTD: Major ad-tech casualty The Trade Desk is down roughly 27% to 30% premarket following weaker-than-expected revenue and guidance. Market impact: Continues the severe earnings reset in high-multiple ad-tech. Potential sympathy pressure remains on digital advertising names. Trading read: Weak rebounds that fail VWAP remain fadeable.

• 
  Celene is in Seattle! (@toasterlighting) reported 32 minutes ago

  @levelsio Yeah, this happened to me too. Annoyingly, meta's web-indexer gets classified as a Search Bot, so cloudflare lets it through a lot despite how rude it is

• 
  Madibo Fred Zadoch (@FredZadoch) reported 39 minutes ago

  Cloudflare unveils an AI-first browser. The company introduced Kitesurf, a web browser designed to help AI agents perform tasks on users' behalf, signaling a move toward more AI-driven browsing.

• 
  Matthew MacKinnon (@MatthewMac69462) reported 43 minutes ago

  @timo_rf One slight issue I have with Cloudflare is that computer-use agents can’t use it unless it’s the focused window on your computer. Not sure if aws is the same way or not.

• 
  biggy (@biggy_kta) reported 48 minutes ago

  Cloudflare shipped programmable wallets for AI agents. Autonomous stablecoin payments for APIs, in production. The payment layer and execution layer are now the same problem. Most infrastructure is going to find that out at the worst possible time.

• 
  Scruffyy90 (@scruffyy90) reported 54 minutes ago

  @Tim_axss I already contacted support. They refused to escalate a ticket. I did some digging and realized that you guys have your Cloudflare settings way too aggressive. The support site is useless as nothing was applicable to my situation

• 
  Debbie O'Brien (@debs_obrien) reported 54 minutes ago

  I still don’t understand @Cloudflare (sorry) but used it twice over the last week for backend services and I have to say the agent experience was amazing cause I still don’t understood Cloudflare but I managed to migrate my Azure service which I didn’t build and was not working anymore and is now using Cloudflare. My playwright demo app is now loading movies but it seems login is not fully working since I deployed. Damn it worked on my machine 🤪. Will fix later. But all this was done over a few hours which is pretty incredible. Thanks @cursor_ai and Grok. Have to say I love how I just typed in the word Cloudflare in the prompt and it suggested adding the Cloudflare plugin and took it from there. That’s great product UX.

• 
  Ravi Ojha (@raviojhax) reported 56 minutes ago

  after our ios app went live, someone has unleashed an AI harness of sorts that's just persistently trying ways to pentest our backend if it was a human, there would've been breaks, but this is just runnign 24x7 since last 3 days the worst part is, we have so many useless entries in db now although, credit where its due, it's found some good stuff that I'll be rate limiting on that most seCuRiTy rEseArCheRs fail to figure out also, one more thing, it is able to pass cloudflare turnstile (although, it does fail like 60-70% of the time)

• 
  BSCN (@BSCNews) reported 59 minutes ago

  Earnings are in: MARA, CleanSpark, and Cloudflare report MARA (@MARA) swung to a $611M Q2 net loss as a $343M Bitcoin markdown hit the books, with revenue down 27% to $174.9M. The miner sold nearly every coin it produced and is leaning into AI infrastructure, adding rights to a 2 GW Texas site. CleanSpark (@CleanSpark_Inc) posted a $240M quarterly loss of its own, with mining revenue sliding 30.5% to $138M, and is steering toward data centers via its 20-year, $6.6B Sandersville lease with an investment-grade tenant. Cloudflare (@Cloudflare) bucked the trend, accelerating to 36% growth at $696.1M in revenue as CEO Matthew Prince pitched the company as the "payment rails for the Agentic Internet," where agent commerce and crypto payments converge.

• 
  Alex MacGregor (@alexmacgregor__) reported 1 hour ago

  @levelsio Open web is 100% dying, I remember the Cloudflare founder saying most of the traffic is accruing to a handful of companies now and that’s going to get worse! No easy fix especially when incentives are stacked against content creation outside of the big platforms.