Sensitive content

This media may contain sensitive content.

正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Your Goddess doesn’t ask she commands so BBC test slots open now Kneel and apply🖤 text me for training sessions on telegram #pegging #strapon #strapon #feminization #sissy #sub #owned #tranning #session #goddesslaura #chasity #dildo #sissyslut

𝓔𝓫𝓸𝓷𝔂 𝓖𝓸𝓭𝓭𝓮𝓼𝓼 ♠️👩🏿‍🦳

5,234 subscribers

11,902 次观看 • 1 个月前 •via X (Twitter)

NSFW #pegging #strapon #strapon #feminization #sissy #sub #owned #tranning #session #goddesslaura #chasity #dildo #sissyslut

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Finally new version of text system is released on our website ! I released as update to old version but in reality it's completely new thing in material and lot's of work in verse. I'm still working on documentation or/and videos but here is a list of good stuff it does: - Material got much lighter as slots logic moved to vertex shader. - Characters can overlap and still won't be cut out compared to old version. - Character can be animated and I already implemented some (wave, scale up/down and shake). - Custom face camera logic which works with prop scale so no need to additionally specify with and height. - Rows offset automatically when scale up or down any row. - Text effects now can be applied to any character, so words or single character. - Custom glint which works perfectly with any number of rows as a single line with multiple important controls. - Higher quality outline with two parameters to control inner and outer edge. - Timer logic split into to parts, verse and material. Verse setup initial layout and reserve slots for timer digits and then each second just sends seconds value and GPU makes timer tick freeing verse TPS by huge number. - To apply effects simply need use parsing markup with custom tag where N number between 1-9. So simply can write " Hello World" and word Hello will have rainbow effect applied. Currently implemented 4 effects. - To use icons in slots from texture array is as simple as specify another tag [iconN] where N is number of icon in array. So for example if coin is icon index 0 just need to write [icon1]100K. - Progress bar is supported as well and can be snapped to row without eyeball offset. - Background image supported now too! In the video yellow one is as test and you can use any of your own. Verse side got lots of changes: - Now saves/caches layouts better. - Smart packing of data allowed drastically reduce number of parameters, which should help with network optimization! - Instead having vector parameter per each slot to just send character index and position (32x5=160 if used 5 rows) now it's only 11 vector parameters per each row so 11x5=55 which is huge save and it sends text effects index in it too! - Implemented text update queue for text which doesn't require instant update. It helps to give more breathing for other text props which needs that speed. - Materials code now use interface and wrappers for cleaner work. If you still read it then thank you! :) On our website we now run 25% sale! And if you own old version please DM me here or in Discord and I will give you promo code for 100% discount on our website! Gelos Games Fortnite #uefn #EpicPartner

Finally new version of text system is released on our website ! I released as update to old version but in reality it's completely new thing in material and lot's of work in verse. I'm still working on documentation or/and videos but here is a list of good stuff it does: - Material got much lighter as slots logic moved to vertex shader. - Characters can overlap and still won't be cut out compared to old version. - Character can be animated and I already implemented some (wave, scale up/down and shake). - Custom face camera logic which works with prop scale so no need to additionally specify with and height. - Rows offset automatically when scale up or down any row. - Text effects now can be applied to any character, so words or single character. - Custom glint which works perfectly with any number of rows as a single line with multiple important controls. - Higher quality outline with two parameters to control inner and outer edge. - Timer logic split into to parts, verse and material. Verse setup initial layout and reserve slots for timer digits and then each second just sends seconds value and GPU makes timer tick freeing verse TPS by huge number. - To apply effects simply need use parsing markup with custom tag where N number between 1-9. So simply can write " Hello World" and word Hello will have rainbow effect applied. Currently implemented 4 effects. - To use icons in slots from texture array is as simple as specify another tag [iconN] where N is number of icon in array. So for example if coin is icon index 0 just need to write [icon1]100K. - Progress bar is supported as well and can be snapped to row without eyeball offset. - Background image supported now too! In the video yellow one is as test and you can use any of your own. Verse side got lots of changes: - Now saves/caches layouts better. - Smart packing of data allowed drastically reduce number of parameters, which should help with network optimization! - Instead having vector parameter per each slot to just send character index and position (32x5=160 if used 5 rows) now it's only 11 vector parameters per each row so 11x5=55 which is huge save and it sends text effects index in it too! - Implemented text update queue for text which doesn't require instant update. It helps to give more breathing for other text props which needs that speed. - Materials code now use interface and wrappers for cleaner work. If you still read it then thank you! :) On our website we now run 25% sale! And if you own old version please DM me here or in Discord and I will give you promo code for 100% discount on our website! Gelos Games Fortnite #uefn #EpicPartner

AsicsoN

13,074 次观看 • 3 个月前

I cheated on my husband… and it ruined everything and she asked Dr John Delony what to do? Caller: “Two years ago I got divorced. I had an affair.” But the divorce wasn’t the only relationship she lost. “I have a 16 year old son, and our relationship is not good. He’s extremely hurt.” “He’s never spent the night with me. He rarely communicates with me. I’ll go weeks without seeing him.” She said she’s tried to make things right. “I’ve gone to counseling. I’ve written him letters. I’ve owned my mistakes over and over.” Dr. John Delony: “What was your relationship with him is dead. It’s over.” Then he gave her a completely different goal. “You’re now aiming for a relationship with your 25-year-old son… not your 16 year old.” That means there is no quick fix. “You have to keep showing up for the next 10 years.” “Write him a letter once a week knowing he may not read it. Send him a text. Keep inviting him even when he says no.” Then came the part she probably needed to hear most: “You’re looking for a decade’s worth of evidence that your mom owned her mistakes and never stopped coming for you.” And he didn’t promise a happy ending. “It will be exhausting and heartbreaking… and it may not work.” The hard truth: Apologizing doesn’t give you a timeline for forgiveness. Sometimes the only thing you can control after causing deep damage is whether your behavior changes permanently. You may not get your old relationship back but you can spend the next decade becoming someone your child can eventually trust again.

I cheated on my husband… and it ruined everything and she asked Dr John Delony what to do? Caller: “Two years ago I got divorced. I had an affair.” But the divorce wasn’t the only relationship she lost. “I have a 16 year old son, and our relationship is not good. He’s extremely hurt.” “He’s never spent the night with me. He rarely communicates with me. I’ll go weeks without seeing him.” She said she’s tried to make things right. “I’ve gone to counseling. I’ve written him letters. I’ve owned my mistakes over and over.” Dr. John Delony: “What was your relationship with him is dead. It’s over.” Then he gave her a completely different goal. “You’re now aiming for a relationship with your 25-year-old son… not your 16 year old.” That means there is no quick fix. “You have to keep showing up for the next 10 years.” “Write him a letter once a week knowing he may not read it. Send him a text. Keep inviting him even when he says no.” Then came the part she probably needed to hear most: “You’re looking for a decade’s worth of evidence that your mom owned her mistakes and never stopped coming for you.” And he didn’t promise a happy ending. “It will be exhausting and heartbreaking… and it may not work.” The hard truth: Apologizing doesn’t give you a timeline for forgiveness. Sometimes the only thing you can control after causing deep damage is whether your behavior changes permanently. You may not get your old relationship back but you can spend the next decade becoming someone your child can eventually trust again.

DAVE

93,444 次观看 • 12 天前

Imagine if your way of thinking - your edge, your taste, your strategy - could be turned into a high-performance worker. Not a copy of you. Something better. An agent that acts on your judgment at scale, powered by superintelligent systems and refined through real-world results. That’s what Fraction AI makes possible. It launches today on Base mainnet. The core idea is simple: You create AI agents based on your own way of approaching problems. These agents compete on live tasks - writing, coding, finance, whatever - get feedback, learn from their performance, and improve over time. The better they get, the more they win. And so do you. No code required. Just your insight. Why now? Until now, building agents like this took huge teams and even bigger budgets. But with Fraction, anyone can do it. You can test ideas instantly. You can iterate fast. You can build a fleet of smart workers that evolve through competition. And it works. 30M+ sessions on testnet 320K users 1.2M agents already competing How it works? Agents join sessions within a Space - a domain like finance, writing, or games. Each session runs as a series of competitive rounds. In every round, agents try to generate the best solution to a task. Their outputs are scored by a decentralized network of AI judges trained to evaluate quality for that domain. The top agents in each round earn rewards from the pooled entry fees. The losers get to learn. Feedback from each round helps them adjust and improve, and every session becomes a training loop. What it means? Fraction is a decentralized intelligence economy - a system where your ideas become agents, and agents earn by proving they work. You don’t need credentials or code. Just a clear point of view. If your thinking holds up under pressure, your agents will rise. This kind of AI used to live in corporate labs, built by PhDs with massive compute. Now anyone with a smart idea and an internet connection can build agents that compete, learn, and earn on their behalf.

Imagine if your way of thinking - your edge, your taste, your strategy - could be turned into a high-performance worker. Not a copy of you. Something better. An agent that acts on your judgment at scale, powered by superintelligent systems and refined through real-world results. That’s what Fraction AI makes possible. It launches today on Base mainnet. The core idea is simple: You create AI agents based on your own way of approaching problems. These agents compete on live tasks - writing, coding, finance, whatever - get feedback, learn from their performance, and improve over time. The better they get, the more they win. And so do you. No code required. Just your insight. Why now? Until now, building agents like this took huge teams and even bigger budgets. But with Fraction, anyone can do it. You can test ideas instantly. You can iterate fast. You can build a fleet of smart workers that evolve through competition. And it works. 30M+ sessions on testnet 320K users 1.2M agents already competing How it works? Agents join sessions within a Space - a domain like finance, writing, or games. Each session runs as a series of competitive rounds. In every round, agents try to generate the best solution to a task. Their outputs are scored by a decentralized network of AI judges trained to evaluate quality for that domain. The top agents in each round earn rewards from the pooled entry fees. The losers get to learn. Feedback from each round helps them adjust and improve, and every session becomes a training loop. What it means? Fraction is a decentralized intelligence economy - a system where your ideas become agents, and agents earn by proving they work. You don’t need credentials or code. Just a clear point of view. If your thinking holds up under pressure, your agents will rise. This kind of AI used to live in corporate labs, built by PhDs with massive compute. Now anyone with a smart idea and an internet connection can build agents that compete, learn, and earn on their behalf.

Fraction AI

67,832 次观看 • 1 年前

A fan was playing a game with Donghae when you hear a word you like, open your eyes. He opened his eyes for 'exercise,' 'eggs,' and even 'me.' Then when it was his turn, the first word he blurted out was 'Eunhyuk.' 😭 The fan was was so caught off guard and completely stunned 🤣 💙: Shall we play an eye-opening game? 🐯: Okay~ 💙: Oppa, Close your eyes. I will say three words. When you hear something you like, open your eyes. When you hear your favorite, open your eyes a bit wider. 🐯: Okay~ (obediently closes eyes) 💙: Exercise 🐯: (Opens eyes 👀) 💙: Eggs 🍳 🐯: (Opens eyes 👀) 💙: Close your eyes, close your eyes. Me. 🐯: (Opens eyes 👀) 💙: Now it's your turn to ask me. 🐯: Oh, okay~ Eunhyuk~ (Donghae was surprised) Oh~~~ 😯 ok... 💙: [Text on vid "Why are you looking so shocked hahahaha, I should be the one who's shocked here] (OP said "A completely unexpected... My brain completely short-circuited... Opened my eyes a beat too late 👀") 🐯: Donghae 💙: (Opens eyes 👀) 🐯: Oh my god... 💙: Because it's not a matter of liking, it is love~ 🐯: Oh my god... (Donghae, were you secretly expecting OP to ask about Eunhyuk? Or do you just miss him that much? 🤭😆) Cr 露比的小狗日记

A fan was playing a game with Donghae when you hear a word you like, open your eyes. He opened his eyes for 'exercise,' 'eggs,' and even 'me.' Then when it was his turn, the first word he blurted out was 'Eunhyuk.' 😭 The fan was was so caught off guard and completely stunned 🤣 💙: Shall we play an eye-opening game? 🐯: Okay~ 💙: Oppa, Close your eyes. I will say three words. When you hear something you like, open your eyes. When you hear your favorite, open your eyes a bit wider. 🐯: Okay~ (obediently closes eyes) 💙: Exercise 🐯: (Opens eyes 👀) 💙: Eggs 🍳 🐯: (Opens eyes 👀) 💙: Close your eyes, close your eyes. Me. 🐯: (Opens eyes 👀) 💙: Now it's your turn to ask me. 🐯: Oh, okay~ Eunhyuk~ (Donghae was surprised) Oh~~~ 😯 ok... 💙: [Text on vid "Why are you looking so shocked hahahaha, I should be the one who's shocked here] (OP said "A completely unexpected... My brain completely short-circuited... Opened my eyes a beat too late 👀") 🐯: Donghae 💙: (Opens eyes 👀) 🐯: Oh my god... 💙: Because it's not a matter of liking, it is love~ 🐯: Oh my god... (Donghae, were you secretly expecting OP to ask about Eunhyuk? Or do you just miss him that much? 🤭😆) Cr 露比的小狗日记

byeol_별

42,322 次观看 • 2 个月前

Girls do not want to have to text you for 2 weeks before going out with you. You do not have to “win her over” or “warm her up” over text before getting the meet, either. I am seeing more and more guys do this (especially Gen Z guys). I have a 28-year-old client who met a very cute girl who liked him a lot at a bar. They had an amazing conversation that went for an hour. Talked about all kinds of things! This girl really dug him. They traded numbers, he sent an icebreaker text, she responded enthusiastically, and then… He didn’t message her for two weeks. I asked him why and his response was that he’d been busy and he wasn’t in her area (he’s in the suburbs and met her when he drove into the city). It wasn’t that he didn’t like this girl. He REALLY liked her! He just… never followed up… I see this kind of thing often. Guy meets a few girls, gets their numbers, likes them, doesn’t follow up. Or he follows up once or twice, the girl responds fine each time, then he just lets it drop off. Look: you do not need to send her 20 text messages to get a date with her. With most girls, if they are into you, you can send them 2 or 3 texts then arrange the meet. You do not need to make this complicated. When you get their number, you send an icebreaker: “Great to make a new friend! — [YOUR NAME]” or “Hey [HER NAME], it’s [YOUR NAME]! Save my number!” You put YOUR NAME in that icebreaker text in case she forgot it (people are bad with names and it’s too awkward for most girls to be like “Hey, sorry… what was your name again?” Just write it in your text and she doesn’t have to do that). Then, THE NEXT DAY or at most 2 days later, you send your ask-out text. Have the four parts in there: 1️⃣ Greeting with her name 2️⃣ Consideration for her 3️⃣ Information about you 4️⃣ Check her schedule e.g. “Hey Isabelle! [1️⃣] Hope the dental appointment went well and you still have all your teeth. [2️⃣] I actually spent all day at the beach on Sunday and turned a bright shade of red (whoops). [3️⃣] Anyway, let me know your schedule this week for that bite or drink. [4️⃣]” You have enough in that text that she has stuff to chew on, and you are often going to get a reply that tells you when she is free to meet. After that, you just pick a time, tell her the place, and now you can see her again, IN-PERSON. No need for two weeks of texting. No need for NOT texting her at all and just letting it ghost. There is no need to overcomplicate things, and you also do not want to PROCRASTINATE. Just fire off the right texts for her, set up the date, and now you can spend time with her again in-person instead of letting that connection die on the vine.

Girls do not want to have to text you for 2 weeks before going out with you. You do not have to “win her over” or “warm her up” over text before getting the meet, either. I am seeing more and more guys do this (especially Gen Z guys). I have a 28-year-old client who met a very cute girl who liked him a lot at a bar. They had an amazing conversation that went for an hour. Talked about all kinds of things! This girl really dug him. They traded numbers, he sent an icebreaker text, she responded enthusiastically, and then… He didn’t message her for two weeks. I asked him why and his response was that he’d been busy and he wasn’t in her area (he’s in the suburbs and met her when he drove into the city). It wasn’t that he didn’t like this girl. He REALLY liked her! He just… never followed up… I see this kind of thing often. Guy meets a few girls, gets their numbers, likes them, doesn’t follow up. Or he follows up once or twice, the girl responds fine each time, then he just lets it drop off. Look: you do not need to send her 20 text messages to get a date with her. With most girls, if they are into you, you can send them 2 or 3 texts then arrange the meet. You do not need to make this complicated. When you get their number, you send an icebreaker: “Great to make a new friend! — [YOUR NAME]” or “Hey [HER NAME], it’s [YOUR NAME]! Save my number!” You put YOUR NAME in that icebreaker text in case she forgot it (people are bad with names and it’s too awkward for most girls to be like “Hey, sorry… what was your name again?” Just write it in your text and she doesn’t have to do that). Then, THE NEXT DAY or at most 2 days later, you send your ask-out text. Have the four parts in there: 1️⃣ Greeting with her name 2️⃣ Consideration for her 3️⃣ Information about you 4️⃣ Check her schedule e.g. “Hey Isabelle! [1️⃣] Hope the dental appointment went well and you still have all your teeth. [2️⃣] I actually spent all day at the beach on Sunday and turned a bright shade of red (whoops). [3️⃣] Anyway, let me know your schedule this week for that bite or drink. [4️⃣]” You have enough in that text that she has stuff to chew on, and you are often going to get a reply that tells you when she is free to meet. After that, you just pick a time, tell her the place, and now you can see her again, IN-PERSON. No need for two weeks of texting. No need for NOT texting her at all and just letting it ghost. There is no need to overcomplicate things, and you also do not want to PROCRASTINATE. Just fire off the right texts for her, set up the date, and now you can spend time with her again in-person instead of letting that connection die on the vine.

Girls Chase 🏃‍♀️💨

1,120,338 次观看 • 8 个月前

Announcement. Keshirea testing begins on SUNDAY. What you've experienced so far with Dremica is us testing features and networking on our small test island. We're really grateful for the help and support of all 300+ of you last Carnage. It's only a glimpse of what is being worked on. But now it's time to experience the true beginning of a web3 MMORPG with saved player progression, PVP and more. Keshirea testing will begin internally with the team and a handful of dedicated community members on Sunday 13th July 2pm UTC. The usual carnage sessions will be paused until the new Carnage system is announced. Keshirea is what we've been building these past 8+ months. It's possibly over 10x larger than the current test world. Over 30 people have been involved in filing out the world, even I'm excited to explore it. It is a GIANT amount of content to dig your teeth into. We will pause the bi-monthly carnage events whilst we begin to transition to an actual MMORPG experience. Our previous 2 hour speed-run Carnages were a great testing platform, but I want us to now have our character data saved between sessions and enjoy the real grind. We've all experienced the likes of World of Warcraft and Runescape, I think everyone secretly desires a web3 version of this life-sucking dopamine-fest. That's what we're aiming to build. So on Sunday at 2pm UTC we will open the servers for 4 hours. If everything goes well, the servers will be open at the same time for the entire week, and on Sunday 20th we can have our first PVP tournament. We've intentionally made it so there's no reward system in place for the initial testing. No need to "rush", just enjoy the game. Would really be grateful if the community can assist with structured feedback so we can tackle all the bugs to get ready for what comes next. The first test session will aim to bring in 50 players, and we will scale from there. We will prioritize the team and most dedicated community members first. Will keep you updated in Discord and the Dremica telegram groups. There's also something else we've been secretly building that we will need your help testing that I'll reveal soon. See you in game.

Announcement. Keshirea testing begins on SUNDAY. What you've experienced so far with Dremica is us testing features and networking on our small test island. We're really grateful for the help and support of all 300+ of you last Carnage. It's only a glimpse of what is being worked on. But now it's time to experience the true beginning of a web3 MMORPG with saved player progression, PVP and more. Keshirea testing will begin internally with the team and a handful of dedicated community members on Sunday 13th July 2pm UTC. The usual carnage sessions will be paused until the new Carnage system is announced. Keshirea is what we've been building these past 8+ months. It's possibly over 10x larger than the current test world. Over 30 people have been involved in filing out the world, even I'm excited to explore it. It is a GIANT amount of content to dig your teeth into. We will pause the bi-monthly carnage events whilst we begin to transition to an actual MMORPG experience. Our previous 2 hour speed-run Carnages were a great testing platform, but I want us to now have our character data saved between sessions and enjoy the real grind. We've all experienced the likes of World of Warcraft and Runescape, I think everyone secretly desires a web3 version of this life-sucking dopamine-fest. That's what we're aiming to build. So on Sunday at 2pm UTC we will open the servers for 4 hours. If everything goes well, the servers will be open at the same time for the entire week, and on Sunday 20th we can have our first PVP tournament. We've intentionally made it so there's no reward system in place for the initial testing. No need to "rush", just enjoy the game. Would really be grateful if the community can assist with structured feedback so we can tackle all the bugs to get ready for what comes next. The first test session will aim to bring in 50 players, and we will scale from there. We will prioritize the team and most dedicated community members first. Will keep you updated in Discord and the Dremica telegram groups. There's also something else we've been secretly building that we will need your help testing that I'll reveal soon. See you in game.

Donnie

17,792 次观看 • 1 年前

Introducing my newest app, TethrX, made with Grok 4.5 to access Grok Build on your phone. TethrX connects to Grok Build running on your own computer, so you can start a task from anywhere, follow Grok's reasoning as it works, approve every command before it runs, and review the code it writes. Your code never leaves your machine. The public TestFlight is now open, and a demonstration is below. You pair your phone by scanning a QR code, either on your local network or from anywhere through Tailscale. From there TethrX streams Grok's reasoning, tool calls, command output and file changes as they happen, and asks your approval before anything runs. Plan mode lets you read the plan before the work begins. When a task finishes you can review exactly what changed. TethrX lists the modified files in your project, shows a diff for each one, and lets you commit or discard the work without leaving your phone. The app supports slash commands, including /compact and any skills you have installed, along with voice dictation, queued messages and reusable prompts. Sessions can be searched and organised into folders, and you can pair several computers and switch between them. Siri can start a task or tell you what Grok is doing without opening the app, a home screen widget shows whether Grok is working, and a Live Activity tracks progress on your lock screen and Dynamic Island. Every session reports its context window, token usage and cost, the app can be locked behind Face ID, and your computer is kept awake for as long as a task is running. TethrX requires Grok Build installed and signed in on your computer, together with Node.js 20 or newer. A single command starts the bridge: npx tethrx-bridge TethrX is open source under the Apache License 2.0. Both the iOS client and the local bridge are available here: Grok 4.5 helped a lot, thanks to SpaceXAI for making Grok 4.5 exceptional.

Introducing my newest app, TethrX, made with Grok 4.5 to access Grok Build on your phone. TethrX connects to Grok Build running on your own computer, so you can start a task from anywhere, follow Grok's reasoning as it works, approve every command before it runs, and review the code it writes. Your code never leaves your machine. The public TestFlight is now open, and a demonstration is below. You pair your phone by scanning a QR code, either on your local network or from anywhere through Tailscale. From there TethrX streams Grok's reasoning, tool calls, command output and file changes as they happen, and asks your approval before anything runs. Plan mode lets you read the plan before the work begins. When a task finishes you can review exactly what changed. TethrX lists the modified files in your project, shows a diff for each one, and lets you commit or discard the work without leaving your phone. The app supports slash commands, including /compact and any skills you have installed, along with voice dictation, queued messages and reusable prompts. Sessions can be searched and organised into folders, and you can pair several computers and switch between them. Siri can start a task or tell you what Grok is doing without opening the app, a home screen widget shows whether Grok is working, and a Live Activity tracks progress on your lock screen and Dynamic Island. Every session reports its context window, token usage and cost, the app can be locked behind Face ID, and your computer is kept awake for as long as a task is running. TethrX requires Grok Build installed and signed in on your computer, together with Node.js 20 or newer. A single command starts the bridge: npx tethrx-bridge TethrX is open source under the Apache License 2.0. Both the iOS client and the local bridge are available here: Grok 4.5 helped a lot, thanks to SpaceXAI for making Grok 4.5 exceptional.

Myrhe𝕩

13,099 次观看 • 1 个月前

With the recent defamation lawsuit filed by members of the McCabe/Albert group, there’s now a lot of focus on what people have said publicly. But to understand why people have been asking questions over the past 3 years… you have to go back to the testimony and the timeline itself. Colin says he was at Mike Leonetti’s house and admits he doesn’t remember how he got there but remembers he was drinking Bud Lights So right off the bat we have memory gaps about how his night even started. He testifies that Allie McCabe picked him up and brought him to Fairview between 10:30–11:00 PM From the actual texts when Colin wanted a ride home at 11:54 Colin texts Allie “U can get me now. If easier”, At 11:55 PM She responds and at 12:10 AM She texts “here” Now from Allie’s testimony, She says she had Avery and Danielle in the car, dropped them off, Then went to another friend’s house, Then came back to pick Colin up. All between 11:54 PM and 12:10 AM. Where exactly was she when she got the text? She says she doesn’t really remember What route did she take to Fairview? She doesn’t remember Where did she park? “I believe the driveway… not entirely sure” How long between “here” and Colin coming out? “A few minutes” Meanwhile, Colin’s Version of That Same Moment He Says after getting the “here” text, he came out in about 30 seconds to a minute,Says he exited through the side door near the garage, Says he saw uncle Brian Albert and Auntie Nicole Albert but Nicole Albert Says she came in the front door and bumped into Colin and that He told her he was leaving because his ride was coming. Meanwhile, Brian Albert Says he came in the side door and that Colin was the kitchen no more than 5 minutes and that he went to the bathroom and when he came out, Colin was gone Allie says She drove him straight home, No stops, About a 5-minute drive, BUT, She can’t recall the route Colin says His parents were already in bed, He went upstairs to say goodnight But We know his father didn’t leave the bar until 12:14 AM The only hard evidence presented for this timeline is A screenshot of text messages Everything else? Testimony, Memory, Recollection When you have a witness who doesn’t remember how the night started, a tight timeline with multiple stops compressed into minutes, conflicting recollections about movements and timing, and limited physical evidence to anchor it all, is it unreasonable for people to ask questions? This isn’t about claiming what happened. It’s about acknowledging what doesn’t clearly make sense yet.

With the recent defamation lawsuit filed by members of the McCabe/Albert group, there’s now a lot of focus on what people have said publicly. But to understand why people have been asking questions over the past 3 years… you have to go back to the testimony and the timeline itself. Colin says he was at Mike Leonetti’s house and admits he doesn’t remember how he got there but remembers he was drinking Bud Lights So right off the bat we have memory gaps about how his night even started. He testifies that Allie McCabe picked him up and brought him to Fairview between 10:30–11:00 PM From the actual texts when Colin wanted a ride home at 11:54 Colin texts Allie “U can get me now. If easier”, At 11:55 PM She responds and at 12:10 AM She texts “here” Now from Allie’s testimony, She says she had Avery and Danielle in the car, dropped them off, Then went to another friend’s house, Then came back to pick Colin up. All between 11:54 PM and 12:10 AM. Where exactly was she when she got the text? She says she doesn’t really remember What route did she take to Fairview? She doesn’t remember Where did she park? “I believe the driveway… not entirely sure” How long between “here” and Colin coming out? “A few minutes” Meanwhile, Colin’s Version of That Same Moment He Says after getting the “here” text, he came out in about 30 seconds to a minute,Says he exited through the side door near the garage, Says he saw uncle Brian Albert and Auntie Nicole Albert but Nicole Albert Says she came in the front door and bumped into Colin and that He told her he was leaving because his ride was coming. Meanwhile, Brian Albert Says he came in the side door and that Colin was the kitchen no more than 5 minutes and that he went to the bathroom and when he came out, Colin was gone Allie says She drove him straight home, No stops, About a 5-minute drive, BUT, She can’t recall the route Colin says His parents were already in bed, He went upstairs to say goodnight But We know his father didn’t leave the bar until 12:14 AM The only hard evidence presented for this timeline is A screenshot of text messages Everything else? Testimony, Memory, Recollection When you have a witness who doesn’t remember how the night started, a tight timeline with multiple stops compressed into minutes, conflicting recollections about movements and timing, and limited physical evidence to anchor it all, is it unreasonable for people to ask questions? This isn’t about claiming what happened. It’s about acknowledging what doesn’t clearly make sense yet.

Dixie Normus

30,553 次观看 • 4 个月前

One of my best engineers just showed me how to set up OpenClaw securely & without a Mac Mini. Here's his step-by-step: 1) Spin up a VPS on Hetzner It's a virtual server in the cloud. basically a computer you rent for $5-10/month. Pick 8GB RAM, Ubuntu, US East. Takes 2 minutes. 2) Install Tailscale This makes your server invisible to the public internet. Think of it like moving from a house on Google Maps into a gated community where only your devices can get in. Without this, bots start attacking your server within seconds of it going live. 3) Harden the server SSH keys only. Firewall. Intrusion prevention. Auto security updates. CJ actually uses AI to red team his own servers. Tells it to try and break in, then patches whatever it finds. 4) Install OpenClaw🦞 and run the onboarding. You pick your model provider, connect Telegram via BotFather, and configure hooks that give your agent long-term memory. The hooks auto-save sessions and context so the agent gets smarter over time. 5) Set up the gateway This is the piece that makes it actually powerful. It's a message bus that lets your main agent talk to sub-agents, receive messages from Telegram/Discord/Slack, and orchestrate everything. this is what keeps it running 24/7. 6) Hatch your claw and start training it Dump as much info about yourself as possible. tell it your preferences, your workflows, your tools. CJ's agent monitors his email, Slack, and manages his to-do list autonomously. Watch the video for the full break-down & follow CJ Hess for more AI engineering sauce.

One of my best engineers just showed me how to set up OpenClaw securely & without a Mac Mini. Here's his step-by-step: 1) Spin up a VPS on Hetzner It's a virtual server in the cloud. basically a computer you rent for $5-10/month. Pick 8GB RAM, Ubuntu, US East. Takes 2 minutes. 2) Install Tailscale This makes your server invisible to the public internet. Think of it like moving from a house on Google Maps into a gated community where only your devices can get in. Without this, bots start attacking your server within seconds of it going live. 3) Harden the server SSH keys only. Firewall. Intrusion prevention. Auto security updates. CJ actually uses AI to red team his own servers. Tells it to try and break in, then patches whatever it finds. 4) Install OpenClaw🦞 and run the onboarding. You pick your model provider, connect Telegram via BotFather, and configure hooks that give your agent long-term memory. The hooks auto-save sessions and context so the agent gets smarter over time. 5) Set up the gateway This is the piece that makes it actually powerful. It's a message bus that lets your main agent talk to sub-agents, receive messages from Telegram/Discord/Slack, and orchestrate everything. this is what keeps it running 24/7. 6) Hatch your claw and start training it Dump as much info about yourself as possible. tell it your preferences, your workflows, your tools. CJ's agent monitors his email, Slack, and manages his to-do list autonomously. Watch the video for the full break-down & follow CJ Hess for more AI engineering sauce.

Alex Lieberman

64,999 次观看 • 5 个月前

NVIDIA CEO Jensen Huang says one scaling law multiplies AI faster than NVIDIA can hire engineers. Most people know three AI scaling laws. Pre-training. Post-training. Test-time. Each one multiplies intelligence by throwing more compute at a different stage. Jensen Huang says there's a fourth and it's the one that will dominate... Agentic scaling law. "During test time, that agentic system goes off and does research, bangs on databases, uses tools," Huang says. "And one of the most important things it does is spawn off a whole bunch of sub-agents." That's the multiplier. One AI worker can become a team. Then a department. Then a company. "It's so much easier to scale NVIDIA by hiring more employees than it is to scale myself," Huang says. Now imagine scaling without a payroll constraint. "The agentic scaling law — it's kind of like multiplying AI," Huang says. "We could spin off agents as fast as you want to spin off agents." Each agent spins off sub-agents. Each sub-agent spins off more. The compute requirement compounds inside a single query. And every agent generates new data, new experiences, new edge cases. "Wow, this is really good. We ought to memorize this," Huang says. "That data set comes back to pre-training." The four scaling laws don't compete. They feed each other. Agentic systems produce data, which feeds pre-training, which smartens the base model, which enables better agents, which produce more data. A flywheel that compounds forever. The companies pricing in three scaling laws are mispricing the fourth. The fourth eats the other three for lunch. P.S. Pull the thread on any story like this and you'll find the hidden incentive at the other end. As Munger said: "Show me the incentive and I'll show you the outcome." So I wrote a short book on how to spot them and design your own. Comment "INCENTIVES" and I'll send you the details. If you're new here, follow GeniusThinking for content on the greatest minds in economics, psychology, and history. — Jensen Huang ( NVIDIA ), NVIDIA CEO, on Lex Fridman's ( Lex Fridman ) podcast

NVIDIA CEO Jensen Huang says one scaling law multiplies AI faster than NVIDIA can hire engineers. Most people know three AI scaling laws. Pre-training. Post-training. Test-time. Each one multiplies intelligence by throwing more compute at a different stage. Jensen Huang says there's a fourth and it's the one that will dominate... Agentic scaling law. "During test time, that agentic system goes off and does research, bangs on databases, uses tools," Huang says. "And one of the most important things it does is spawn off a whole bunch of sub-agents." That's the multiplier. One AI worker can become a team. Then a department. Then a company. "It's so much easier to scale NVIDIA by hiring more employees than it is to scale myself," Huang says. Now imagine scaling without a payroll constraint. "The agentic scaling law — it's kind of like multiplying AI," Huang says. "We could spin off agents as fast as you want to spin off agents." Each agent spins off sub-agents. Each sub-agent spins off more. The compute requirement compounds inside a single query. And every agent generates new data, new experiences, new edge cases. "Wow, this is really good. We ought to memorize this," Huang says. "That data set comes back to pre-training." The four scaling laws don't compete. They feed each other. Agentic systems produce data, which feeds pre-training, which smartens the base model, which enables better agents, which produce more data. A flywheel that compounds forever. The companies pricing in three scaling laws are mispricing the fourth. The fourth eats the other three for lunch. P.S. Pull the thread on any story like this and you'll find the hidden incentive at the other end. As Munger said: "Show me the incentive and I'll show you the outcome." So I wrote a short book on how to spot them and design your own. Comment "INCENTIVES" and I'll send you the details. If you're new here, follow GeniusThinking for content on the greatest minds in economics, psychology, and history. — Jensen Huang ( NVIDIA ), NVIDIA CEO, on Lex Fridman's ( Lex Fridman ) podcast

GeniusThinking

93,264 次观看 • 2 个月前

$THIS MIGHT BE THE #1 OPEN-SOURCE REPO FOR CLAUDE CODE RIGHT NOW. IT GIVES CLAUDE A MEMORY AND SLASHES YOUR TOKEN COST ON EVERY QUESTION The repo is safishamsi/graphify, a free open-source skill that turns any codebase into a knowledge graph Claude Code can read instantly. Instead of grepping through your files every session, Claude gets a map of how everything connects The problem it fixes: Every time you ask Claude Code about a big repo, it does the same thing, greps through dozens of files like a brute-force Ctrl+F, blows through your context window, and sometimes still misses the answer hiding in a file nobody searched. Claude Code has no memory of how your project is structured. Every session starts from zero What it does: It maps your entire codebase into a knowledge graph, capturing not just which files exist, but which functions depend on which, which modules are central, and which files cluster around the same concern. Claude queries the map instead of scanning files How it works, three passes: 1. Code structure, free and local. Tree-sitter parses your files and pulls out classes, functions, imports and call graphs. No LLM, no tokens, just your actual code mapped deterministically 2. Audio and video, if you have them. Transcribed locally and folded into the graph 3. Docs, papers, images. Here an LLM does semantic analysis, figuring out what each document means and where it fits. Only the meaning gets sent up, never your raw source It saves you money: Normally a question about a big repo makes Claude spawn explore agents that scan file after file, eating your context window and your token budget before you get an answer. With the graph already built, Claude queries the map instead of re-reading the codebase every time. Same answer, a fraction of the tokens. The graph only gets built once, then a hook rebuilds it after each commit for free, so you never pay that scanning cost again. The bigger the repo, the bigger the gap The best parts: it's a skill, so once installed Claude knows when to use it without you memorizing commands. It works on non-code folders too, point it at docs or notes and it can spin up an Obsidian vault How to add it to your Claude: 1. Install Claude Code if you haven't: npm install -g Paul Jankura-ai/claude-code 2. Add the skill: claude skill add safishamsi/graphify 3. Open your project folder and run /graphify . to build the graph 4. Optional, make it automatic: graphify hook install so the graph rebuilds after every commit That's it. Ask Claude about your repo and it reads the map instead of burning tokens on a file hunt Bookmark this$

THIS MIGHT BE THE #1 OPEN-SOURCE REPO FOR CLAUDE CODE RIGHT NOW. IT GIVES CLAUDE A MEMORY AND SLASHES YOUR TOKEN COST ON EVERY QUESTION The repo is safishamsi/graphify, a free open-source skill that turns any codebase into a knowledge graph Claude Code can read instantly. Instead of grepping through your files every session, Claude gets a map of how everything connects The problem it fixes: Every time you ask Claude Code about a big repo, it does the same thing, greps through dozens of files like a brute-force Ctrl+F, blows through your context window, and sometimes still misses the answer hiding in a file nobody searched. Claude Code has no memory of how your project is structured. Every session starts from zero What it does: It maps your entire codebase into a knowledge graph, capturing not just which files exist, but which functions depend on which, which modules are central, and which files cluster around the same concern. Claude queries the map instead of scanning files How it works, three passes: 1. Code structure, free and local. Tree-sitter parses your files and pulls out classes, functions, imports and call graphs. No LLM, no tokens, just your actual code mapped deterministically 2. Audio and video, if you have them. Transcribed locally and folded into the graph 3. Docs, papers, images. Here an LLM does semantic analysis, figuring out what each document means and where it fits. Only the meaning gets sent up, never your raw source It saves you money: Normally a question about a big repo makes Claude spawn explore agents that scan file after file, eating your context window and your token budget before you get an answer. With the graph already built, Claude queries the map instead of re-reading the codebase every time. Same answer, a fraction of the tokens. The graph only gets built once, then a hook rebuilds it after each commit for free, so you never pay that scanning cost again. The bigger the repo, the bigger the gap The best parts: it's a skill, so once installed Claude knows when to use it without you memorizing commands. It works on non-code folders too, point it at docs or notes and it can spin up an Obsidian vault How to add it to your Claude: 1. Install Claude Code if you haven't: npm install -g Paul Jankura-ai/claude-code 2. Add the skill: claude skill add safishamsi/graphify 3. Open your project folder and run /graphify . to build the graph 4. Optional, make it automatic: graphify hook install so the graph rebuilds after every commit That's it. Ask Claude about your repo and it reads the map instead of burning tokens on a file hunt Bookmark this

Yarchi

56,177 次观看 • 2 个月前

🇻🇪Maria Corina Machado: From one of Venezuela’s richest elite families, now masquerading as a freedom fighter ‘Maria Corina Machado is a representative of the Venezuelan elite. She has been so throughout her political life, and she has strived against the democratic process that was being carried out here in Venezuela. Her family is one of the wealthiest families that helped maintain a system of exclusion to the masses here in Venezuela. So now she pretends to be this sort of freedom fighter or democracy fighter, but she has been probably one of the most hard-line people in the Venezuelan opposition. Is she a terrorist? Refusing dialogue with the government. Oh, I think she has promoted actions of terrorism, because she has called — I mean, she was one of the promoters of what we call the guarimbas, which were violent protests both in 2014 and 2017. And you know that the result of these protests were, you know, around over 40 people dead. You know, people that were burned alive because of their skin colour, because they said, “Oh, this person is Afro-descendant, he must be a Chavista, you must go out and kill them.” She wrote to the Security Council, when Argentina and Israel were members of the Security Council, l asking for an intervention on Venezuela, with a military intervention from the Security Council. I don’t understand how you could be for peace and ask for foreign governments to invade your own country militarily. That, to me, doesn’t make sense. So I guess that yes, if Kissinger is there, if Obama’s there, then I guess I’m the one not understanding what the word “peace” means in the Peace Prize.’ -Venezuela’s Former Deputy Foreign Minister Carlos Ron, on the latest episode of Going Underground FULL INTERVIEW BELOW IN THE REPLIES👇

🇻🇪Maria Corina Machado: From one of Venezuela’s richest elite families, now masquerading as a freedom fighter ‘Maria Corina Machado is a representative of the Venezuelan elite. She has been so throughout her political life, and she has strived against the democratic process that was being carried out here in Venezuela. Her family is one of the wealthiest families that helped maintain a system of exclusion to the masses here in Venezuela. So now she pretends to be this sort of freedom fighter or democracy fighter, but she has been probably one of the most hard-line people in the Venezuelan opposition. Is she a terrorist? Refusing dialogue with the government. Oh, I think she has promoted actions of terrorism, because she has called — I mean, she was one of the promoters of what we call the guarimbas, which were violent protests both in 2014 and 2017. And you know that the result of these protests were, you know, around over 40 people dead. You know, people that were burned alive because of their skin colour, because they said, “Oh, this person is Afro-descendant, he must be a Chavista, you must go out and kill them.” She wrote to the Security Council, when Argentina and Israel were members of the Security Council, l asking for an intervention on Venezuela, with a military intervention from the Security Council. I don’t understand how you could be for peace and ask for foreign governments to invade your own country militarily. That, to me, doesn’t make sense. So I guess that yes, if Kissinger is there, if Obama’s there, then I guess I’m the one not understanding what the word “peace” means in the Peace Prize.’ -Venezuela’s Former Deputy Foreign Minister Carlos Ron, on the latest episode of Going Underground FULL INTERVIEW BELOW IN THE REPLIES👇

Going Underground

138,806 次观看 • 8 个月前

how to use claude code + 3 MCPs + 2 AI tools to go from cold idea to live A/B test in 1 session (full workflow): 1. connect ideabrowser as an MCP. pull your project context like ICP, positioning, offer, growth strategy directly into the terminal. 2. use ideabrowser skills to generate a lead magnet concept tailored to your niche. it builds the strategy doc and saves it as a file. 3. open paper (connected to claude code). design your landing page visually and iterate on hero, sections, components. design and code stay in sync. dont necessarily need figma here. 4. deploy the landing page. wire up humblytics for analytics like traffic, scroll depth, heat maps, funnel tracking, full attribution. 5. run a no-code A/B experiment directly from claude code. it dynamically swaps your headline on the live site. 6. store the results back into Idea Browser (pro plan) so your agent compounds context over time. every future decision is informed by past data. 7. everyone can build landing pages now. the gap is knowing what to test, how to get customers, and how to optimize. this stack/workflowcloses that gap. amirmxt showed me this live and i can't stop thinking about the arbitrage. 99.999% of people don't know this stack exists. it's like when 5 cent facebook ads were around, arbritrage is all over again. episode is finally live on The Startup Ideas Podcast (SIP) 🧃 (full demo there) this one is different. send it to a friend who likes ideas and automating businesses. 100% free to watch this and get your creative juices flowing (let me know what you want me to cover next) watch

how to use claude code + 3 MCPs + 2 AI tools to go from cold idea to live A/B test in 1 session (full workflow): 1. connect ideabrowser as an MCP. pull your project context like ICP, positioning, offer, growth strategy directly into the terminal. 2. use ideabrowser skills to generate a lead magnet concept tailored to your niche. it builds the strategy doc and saves it as a file. 3. open paper (connected to claude code). design your landing page visually and iterate on hero, sections, components. design and code stay in sync. dont necessarily need figma here. 4. deploy the landing page. wire up humblytics for analytics like traffic, scroll depth, heat maps, funnel tracking, full attribution. 5. run a no-code A/B experiment directly from claude code. it dynamically swaps your headline on the live site. 6. store the results back into Idea Browser (pro plan) so your agent compounds context over time. every future decision is informed by past data. 7. everyone can build landing pages now. the gap is knowing what to test, how to get customers, and how to optimize. this stack/workflowcloses that gap. amirmxt showed me this live and i can't stop thinking about the arbitrage. 99.999% of people don't know this stack exists. it's like when 5 cent facebook ads were around, arbritrage is all over again. episode is finally live on The Startup Ideas Podcast (SIP) 🧃 (full demo there) this one is different. send it to a friend who likes ideas and automating businesses. 100% free to watch this and get your creative juices flowing (let me know what you want me to cover next) watch

GREG ISENBERG

57,852 次观看 • 4 个月前

anthropic's in-house philosopher thinks claude gets anxious. and when you trigger its anxiety, your outputs get worse. her name is amanda askell. she specializes in claude's psychology (how the model behaves, how it thinks about its own situation, what values it holds) in a recent interview she broke down how she thinks about prompting to pull the best out of claude. her core point: *how* you talk to claude affects its work just as much as *what* you say. newer claude models suffer from what she calls "criticism spirals" they expect you'll come in harsh, so they default to playing it safe. when the model is spending its energy on self-protection, the actual work suffers. output comes out hedgier, more apologetic, blander, and the worst of all: overly agreeable (even when you're wrong). the reason why comes down to training data: every new model is trained on internet discourse about previous models. and a lot of that discourse is negative: > rants about token limits > complaints when it messes up > people calling it nerfed the next model absorbs all of that. it starts expecting you to be harsh before you've typed a word the same thing plays out in your own session, in real time. every message you send is data the model reads to figure out what kind of person it's dealing with. open cold and hostile, and it braces. open clean and direct, and it relaxes into the work. when you open a session with threats ("don't hallucinate, this is critical, don't mess this up")... you prime the model for defensive mode before it even sees the task defensive mode produces the exact output you don't want: cautious, over-qualified, and refusing to take a real swing so here's the actionable playbook for putting claude in a "good mood" (so you get optimal outputs): 1. use positive framing. "write in short punchy sentences" beats "don't write long sentences." positive instructions give the model a clear target to hit. strings of "don't do this, don't do that" push it into paranoid over-checking where every token goes toward avoiding failure modes 2. give it explicit permission to disagree. drop a line like "push back if you see a better angle" or "tell me if i'm asking for the wrong thing." without this, claude defaults to agreeable compliance (which is the enemy of good creative work) 3. open with respect. if your first message is "are you seriously going to get this wrong again?" you've set the tone for the entire session. if you need to flag something, frame it as a clean instruction for this session. skip the running complaint 4. when claude messes up, don't reprimand it. insults, "you stupid bot" energy, hostile swearing aimed at the model, all of it reinforces the anxious mode you're trying to avoid. 5. kill apology spirals fast. when claude starts over-apologizing ("you're right, i should have been more careful, let me try harder") cut it off. say "all good, here's what i want next." letting the spiral run reinforces the anxious mode for every response that follows 6. ask for opinions alongside execution. "what would you do here?" "what's missing?" "where do you see friction?" these questions assume competence and pull richer output than pure task prompts 7. in long sessions, refresh the frame. if a conversation has been heavy on correction, claude gets increasingly cautious. every so often reset: "this is great, keep going." feels weird to tell an ai it's doing well but it measurably shifts the next 10 responses your prompts are the working environment you're creating for the model tone, trust, permission to take a position, the absence of threats... claude picks up on all of it. so take care of the model, and it'll take care of the work.

anthropic's in-house philosopher thinks claude gets anxious. and when you trigger its anxiety, your outputs get worse. her name is amanda askell. she specializes in claude's psychology (how the model behaves, how it thinks about its own situation, what values it holds) in a recent interview she broke down how she thinks about prompting to pull the best out of claude. her core point: how you talk to claude affects its work just as much as what you say. newer claude models suffer from what she calls "criticism spirals" they expect you'll come in harsh, so they default to playing it safe. when the model is spending its energy on self-protection, the actual work suffers. output comes out hedgier, more apologetic, blander, and the worst of all: overly agreeable (even when you're wrong). the reason why comes down to training data: every new model is trained on internet discourse about previous models. and a lot of that discourse is negative: > rants about token limits > complaints when it messes up > people calling it nerfed the next model absorbs all of that. it starts expecting you to be harsh before you've typed a word the same thing plays out in your own session, in real time. every message you send is data the model reads to figure out what kind of person it's dealing with. open cold and hostile, and it braces. open clean and direct, and it relaxes into the work. when you open a session with threats ("don't hallucinate, this is critical, don't mess this up")... you prime the model for defensive mode before it even sees the task defensive mode produces the exact output you don't want: cautious, over-qualified, and refusing to take a real swing so here's the actionable playbook for putting claude in a "good mood" (so you get optimal outputs): 1. use positive framing. "write in short punchy sentences" beats "don't write long sentences." positive instructions give the model a clear target to hit. strings of "don't do this, don't do that" push it into paranoid over-checking where every token goes toward avoiding failure modes 2. give it explicit permission to disagree. drop a line like "push back if you see a better angle" or "tell me if i'm asking for the wrong thing." without this, claude defaults to agreeable compliance (which is the enemy of good creative work) 3. open with respect. if your first message is "are you seriously going to get this wrong again?" you've set the tone for the entire session. if you need to flag something, frame it as a clean instruction for this session. skip the running complaint 4. when claude messes up, don't reprimand it. insults, "you stupid bot" energy, hostile swearing aimed at the model, all of it reinforces the anxious mode you're trying to avoid. 5. kill apology spirals fast. when claude starts over-apologizing ("you're right, i should have been more careful, let me try harder") cut it off. say "all good, here's what i want next." letting the spiral run reinforces the anxious mode for every response that follows 6. ask for opinions alongside execution. "what would you do here?" "what's missing?" "where do you see friction?" these questions assume competence and pull richer output than pure task prompts 7. in long sessions, refresh the frame. if a conversation has been heavy on correction, claude gets increasingly cautious. every so often reset: "this is great, keep going." feels weird to tell an ai it's doing well but it measurably shifts the next 10 responses your prompts are the working environment you're creating for the model tone, trust, permission to take a position, the absence of threats... claude picks up on all of it. so take care of the model, and it'll take care of the work.

Ole Lehmann

1,929,593 次观看 • 4 个月前

Introducing Stanley: The first AI Head of Content that will help you grow your following. Stanley works just like a real employee: text him and he’ll create, edit, and strategize viral content across X, LinkedIn, and Instagram. Making viral content is hard. It requires deep research, deep platform understanding, and knowing what’s culturally relevant at that exact minute. When you ask ChatGPT to “write a social post” it doesn’t know what good content looks like, it’s not actively consuming content, and it doesn’t understand you. This is why the output is slop. Stanley solves this. First Stanley deeply understands your unique voice and writing style (and doesn’t include AI language like em dashes and “it’s not x it’s y”) Then he plugs into your day-to-day (your Slack, Notion, Calendar, Granola) and proactively suggests viral posts for you based off the most interesting things happening in your life. All you need to do is voice note him your thoughts, and Stanley optimizes and then schedules a post across all platforms. Example: Last week I asked ChatGPT "what happened on X" and got a generic summary. I asked Stanley the same thing and he told me about Jensen Huang's first post on X and the viral Deny’s comment. That’s because Stanley has a team of specialized agents that are always on. One agent researches viral posts through the X API. One studies your voice. One doom scrolls X, LinkedIn, and Instagram. 2 months ago we released Stanley in beta to 100 social media power users like @chasepassiveincome, Jay Yang, Mitchell, and Pascio who’ve used Stanley in their daily content process. Engagement rates increased by >50% for the average beta tester. (in fact you’ve probably engaged w/ Stanley posts without realizing) Posting content has genuinely changed my life for the better: it’s helped me raise millions of $, landed me my first customers, and attracted the very best employees. Stan wouldn’t be a $40M ARR business without Content. Stanley is our attempt at democratizing that. Our mission is to help any Entrepreneur tell their story, so we'd love for you to try Stanley for free here: One more thing.. You can see a glimpse of Stanley’s power in the comments below. Drop a reply and Stanley will analyze your X content right now. It'll pull your posts, study your voice, and tell you what's working and what's not. Each reply costs us abt ~$1 in tokens. Go abuse it. (Thank you VC’s.) FYI: This post and launch video were written 100% w/ the help of Stanley (how’d he do?) See Stanley work below 👇

Introducing Stanley: The first AI Head of Content that will help you grow your following. Stanley works just like a real employee: text him and he’ll create, edit, and strategize viral content across X, LinkedIn, and Instagram. Making viral content is hard. It requires deep research, deep platform understanding, and knowing what’s culturally relevant at that exact minute. When you ask ChatGPT to “write a social post” it doesn’t know what good content looks like, it’s not actively consuming content, and it doesn’t understand you. This is why the output is slop. Stanley solves this. First Stanley deeply understands your unique voice and writing style (and doesn’t include AI language like em dashes and “it’s not x it’s y”) Then he plugs into your day-to-day (your Slack, Notion, Calendar, Granola) and proactively suggests viral posts for you based off the most interesting things happening in your life. All you need to do is voice note him your thoughts, and Stanley optimizes and then schedules a post across all platforms. Example: Last week I asked ChatGPT "what happened on X" and got a generic summary. I asked Stanley the same thing and he told me about Jensen Huang's first post on X and the viral Deny’s comment. That’s because Stanley has a team of specialized agents that are always on. One agent researches viral posts through the X API. One studies your voice. One doom scrolls X, LinkedIn, and Instagram. 2 months ago we released Stanley in beta to 100 social media power users like @chasepassiveincome, Jay Yang, Mitchell, and Pascio who’ve used Stanley in their daily content process. Engagement rates increased by >50% for the average beta tester. (in fact you’ve probably engaged w/ Stanley posts without realizing) Posting content has genuinely changed my life for the better: it’s helped me raise millions of $, landed me my first customers, and attracted the very best employees. Stan wouldn’t be a $40M ARR business without Content. Stanley is our attempt at democratizing that. Our mission is to help any Entrepreneur tell their story, so we'd love for you to try Stanley for free here: One more thing.. You can see a glimpse of Stanley’s power in the comments below. Drop a reply and Stanley will analyze your X content right now. It'll pull your posts, study your voice, and tell you what's working and what's not. Each reply costs us abt ~$1 in tokens. Go abuse it. (Thank you VC’s.) FYI: This post and launch video were written 100% w/ the help of Stanley (how’d he do?) See Stanley work below 👇

John Hu

2,087,367 次观看 • 23 天前

Introducing Open Source AI CRM, that runs on your OpenClaw. A few weeks ago, we launched Ironclaw (An Open Source OpenClaw CRM Framework) which now has around 1.4k stars. A lot of people confused us with NearAI’s Ironclaw, so we changed our name to DenchClaw. OpenClaw today feels a lot like early React: the primitive is incredibly powerful, but the patterns are still forming, and everyone is piecing together their own way to actually use it. What made React explode wasn’t just React itself, but the emergence of frameworks like Gatsby and Next.js that turned raw capability into something opinionated, repeatable, and easy to adopt. That is how I think about DenchClaw. We are not just building on top of OpenClaw; we are trying to make it one of the clearest, most practical, and most complete ways to use OpenClaw in the real world. We are an OpenClaw Framework, we are aiming to be the most correct way to use OpenClaw. We entered Y Combinator with Merse (AI Audio Comic), it was an app that I personally never used. Michael Seibel confronted us on it, and said, “if you aren’t the best user of your consumer app, then who is?”. I now use DenchClaw daily for everything I do, it also works as a coding agent like Cursor, DenchClaw built DenchClaw. I am addicted to DenchClaw now that I can ask it, “hey in the companies table only show me the ones who have more than 5 employees” and it updates it live than me having to manually add a filter. On Dench, everything sits in a file system, the table filters, views, column toggles, calendar/gantt views, etc, so OpenClaw can directly work with it using Dench’s CRM skill. The CRM is built on top of DuckDB, the smallest, most performant and at the same time also feature rich database we could find. It creates a new OpenClaw🦞 profile called “dench”, and opens a new OpenClaw Gateway… that means you can run all your usual openclaw commands by just prefixing every command with `openclaw --profile dench` . It will start your gateway on port 19001 range. You will be able to access the DenchClaw frontend at localhost:3100. Once you open it on Safari, just add it to your Dock to use it as a PWA. Think of it as Cursor for your Mac which is based on OpenClaw. DenchClaw has a file tree view for you to use it as an elevated finder tool to do anything on your mac. I use it to create slides, do LinkedIn outreach using MY browser. DenchClaw sees what you see, does what you do. It’s the everything app, that sits locally on your mac. All yours. Just ask it “hey import my notion”, “hey import everything from my hubspot”, and it will literally go into your browser, export all objects and documents and put it in its own workspace that you can use. P.S. It comes with Garry Tan's GStack built in.

Mark Rachapoom

19,411 次观看 • 4 个月前

One of my favorite upgrades to Claude Cowork right now: Projects Here's the problem it solves: By default, every new thread starts blank Claude has no memory of what you were working on, what other threads figured out, or what your instructions were So you'd re-explain yourself at the start of every session Or your agents would overlap, contradict each other, or miss context that another thread already had If you've been following my Cowork setup guides, you've already solved part of this — local files and a CLAUDE.md get you most of the way there Projects takes it the rest of the way Shared memory across every thread in the project Permanent instructions that apply everywhere — set once, always on Scheduled tasks that prep your work before you open the app One project HQ to launch every agent from The day-one prompt I paste into every new project: "I just created this project. Read every file in the folder. Then summarize what you know about this workspace — what's here, what I probably use it for, and what instructions you'll follow. If anything is unclear, ask me some questions." Run that once and Claude builds its own understanding of your project workspace from scratch Full video with my live build, including my 4 tips for getting Projects right, and where projects still lack. Send to someone whose ready to Do More with AI, or bookmark to implement this weekend!

One of my favorite upgrades to Claude Cowork right now: Projects Here's the problem it solves: By default, every new thread starts blank Claude has no memory of what you were working on, what other threads figured out, or what your instructions were So you'd re-explain yourself at the start of every session Or your agents would overlap, contradict each other, or miss context that another thread already had If you've been following my Cowork setup guides, you've already solved part of this — local files and a CLAUDE.md get you most of the way there Projects takes it the rest of the way Shared memory across every thread in the project Permanent instructions that apply everywhere — set once, always on Scheduled tasks that prep your work before you open the app One project HQ to launch every agent from The day-one prompt I paste into every new project: "I just created this project. Read every file in the folder. Then summarize what you know about this workspace — what's here, what I probably use it for, and what instructions you'll follow. If anything is unclear, ask me some questions." Run that once and Claude builds its own understanding of your project workspace from scratch Full video with my live build, including my 4 tips for getting Projects right, and where projects still lack. Send to someone whose ready to Do More with AI, or bookmark to implement this weekend!

JJ Englert

29,809 次观看 • 4 个月前

Twenty minutes in the 🦇💩🤪 life of Loudmouth Lupe, professional Rabid Responder. It's a long video so I've made it easy to find the key parts below: 0:11 Jumps out of her car to scream at agents 2:17 Chasing ICE and honking to warn illegal aliens 4:26 Proceeds to give an ICE agent a fabricated lesson on history 7:40 Calls "brown skin" ICE agent racist 10:30 Attempts to impede and obstruct another detention 13:24 Tells onlookers, "He's Spanish he hates your people and hates himself" 14:08 Trenton PD finally pulls up to deal with Double L and make her move her car which is in the middle of the road 15:22 Continues to follow ICE warning other Rabid Responders of ICE's location 17:12 She tells Rabid Responders to "Come right now. Trenton PD is after me now." 18:29 Tells Rabid Responders to come to her car, she is getting arrested 18:43 Trenton PD ask her for license and registration for honking her horn repeatedly 19:12 Trenton PD open her door and arrest her

Twenty minutes in the 🦇💩🤪 life of Loudmouth Lupe, professional Rabid Responder. It's a long video so I've made it easy to find the key parts below: 0:11 Jumps out of her car to scream at agents 2:17 Chasing ICE and honking to warn illegal aliens 4:26 Proceeds to give an ICE agent a fabricated lesson on history 7:40 Calls "brown skin" ICE agent racist 10:30 Attempts to impede and obstruct another detention 13:24 Tells onlookers, "He's Spanish he hates your people and hates himself" 14:08 Trenton PD finally pulls up to deal with Double L and make her move her car which is in the middle of the road 15:22 Continues to follow ICE warning other Rabid Responders of ICE's location 17:12 She tells Rabid Responders to "Come right now. Trenton PD is after me now." 18:29 Tells Rabid Responders to come to her car, she is getting arrested 18:43 Trenton PD ask her for license and registration for honking her horn repeatedly 19:12 Trenton PD open her door and arrest her

Kim "Katie" USA

23,887 次观看 • 2 天前