Introducing Firecrawl Developer Index, an index for supercharging coding agents.
Search 70M+ primary sources including repos, docs, & issues with the highest recall of any coding-specific index.
Ensure agents ship correct, up-to-date code every time!
https://t.co/i3rH3PiKFu
If I was a doctor with a private practice, I would offer to sit with the patient and configure their Claude, ChatGPT, Gemini and Grok with instructions and skills, along with a series of prompts/tasks that the patient can use and that can be shared with the doctor and inform the patient.
LLM , every day at 5pm I want you to ask Joe the following questions , and show him your response. Upon completion , I want you to email me everything in this chat
Every morning I want you to look for new questions I emailed joe that will come from my email with the subject Questions
Rinse and repeat.
You get the point.
Help the patient use AI to help you help the patient.
Thoughts ?
@Skool_biz I would love if we had an option for taking our monthly membership pricing out of our monthly revenue automatically, so we don't have to worry about paying.
@anothercohen I told my Claude to always delegate all of my fable usage any time fable wants to use sub-agents to make them sonnet 5 sub-agents, and that has saved me hella fable 5 usage.
AI gives everyone the tools to create.
That terrifies two groups:
Artists who build their identity
around being able to do what others couldn't.
And people who now have every tool
but still have nothing to say.
How about this Tibo?
How about the fact that when I start a Codex CLI session from Termious, using a remote SSH into my pc from my phone, my Codex CLI sessions show up inside of the ChatGPT app?
I think that's fucking amazing.
You should give us all a reset just for that because why isn't every app like that?
It's so convenient. Shout out to ChatGPT!
“hey, did you use AI to make this? we can’t accept things made using AI.”
my brother in christ, that question makes no sense.
you woke up and unlocked your phone with AI-powered face recognition, checked your email that was filtered for spam by AI, read messages with AI-generated reply suggestions, typed with AI autocorrect and predictive text, listened to music recommended by AI, and then took a photo of your breakfast that was automatically denoised, sharpened, HDR-merged, color-corrected and enhanced by AI.
then you ordered lunch through an app that ranked restaurants with AI, estimated the delivery time with AI, optimized the driver's route with AI, and predicted demand to set prices with... guess what? AI.
you then went to work so you opened google and got AI-generated summaries. you translated a sentence with AI and then you edited a photo and a video with AI object selection and content-aware fill. then you had your first meeting (it's 10am already) and joined a zoom call with AI noise suppression, AI echo cancellation, AI background blur, AI live captions, and... AI speaker framing. you then wrote code (well, not really, you accepted code the AI wrote for you). after all the hard work you decided to then watched netflix and youtube because an AI decided what to show you and you paid for that service using a credit card with AI monitoring your transactions for fraud. then you ordered a Waymo, which is literally a car driven by AI, and walked past an AI-powered security camera.
Did I use AI? Of course, unless you live in the middle of the Bermuda Triangle you are using AI every single day.
But yes, absolutely, let's draw the sacred moral line at some completely random place.
These are called "I'll do it myself" games. Games that are made to basically be the spiritual successor to a dead series. We let these games rock no matter how similar they are to the og
yeah i think this proves that critical thinking isn't a skill that most anti-a.i. people display in these convos because nothing he's saying is making sense.
You talk to computers now, we don't have to type.. nobody writes prompts, let alone for 15 minutes..
You could've connected AI to blender.. u told a mf that don't make 3D "make me 3D"
You could've done it yourself first but needed AI to make you actually execute..
You do what you know how to do..which is use Blender the real way
Nowadays everything boils down to proper communication and the right tool
You had neither in this experiment until you went to blender yourself...and thats why your output sucked
But go be angry caveman and put blame on shiny metal box 😂😂
You can take 15 minutes to write a detailed-enough prompt so that a machine understands how to generate video of what you want, and then spend another 15 minutes course-correcting to make sure it gets the color, lighting and looping effect correctly. Or you can spend 15 minutes just making it. I'm not anti-AI, but it has real limitations and is a shitty tool for a lot of things (still).
in the old ways of ai maybe - but now my agents write the prompts - so i ask in natural language what i want, they build the prompt and fire it off (either mcp or locally in comfyUI), manage the asset in my asset management system. it does help to learn the craft of prompting for reviewing them.
I can't overstate how much ass Blender kicks. I was looking for an old stock video I purchased of a police light, couldn't find it and tried generating one using AI. The results were garbage. So I opened up Blender and whipped this up in 15 mins. Left is AI, right is mine.
im at the park and i just opened a remote session from my phone using @TermiusHQ and i logged into codex. then i went to the chatgpt app and the same conversation was there underneath remote in @ChatGPT@claudeai you need to catch up.
i'm not gonna lie to you. your app is not doing that yet and i don't know why.. I still have to remember to /rc everything before I leave the house even with it auto on..cli sessions not visible on the desktop app..record a skill still not available on windows...
but yeah Chat this is really really cool 😎
i use a combination of this, ELI5, my own brand voice, make each first sentence, bold and i only need to read each first sentence to understand what is happening, always end each message with a "your move" that actually tells me what i need to do next to keep going as a human right.
and explain it to me like i'm five with really really really strict "i hate words" type language thrown in there. you're using way too many unnecessary words and you're showing me too many unnecessary words.
it's working. they've hidden a lot more of their thinking and i see a lot more of what would have been displayed to me kind of in the transparent chat when they're talking to themselves.
they then come back to me just with bold line outputs, so i really only got to read the first sentence of whatever they're saying and the last part and i can have a picture of what I need to do next. You have helped a
lot with this so thank you
i use a combination of this, ELI5, my own brand voice, make each first sentence, bold and i only need to read each first sentence to understand what is happening, always end each message with a "your move" that actually tells me what i need to do next to keep going as a human right.
and explain it to me like i'm five with really really really strict "i hate words" type language thrown in there. you're using way too many unnecessary words and you're showing me too many unnecessary words.
it's working. they've hidden a lot more of their thinking and i see a lot more of what would have been displayed to me kind of in the transparent chat when they're talking to themselves.
they then come back to me just with bold line outputs, so i really only got to read the first sentence of whatever they're saying and the last part and i can have a picture of what I need to do next. You have helped a
lot with this so thank you