Show HN: I'm making an AI scraper called FetchFox https://bit.ly/3Tfmrmu

Show HN: I'm making an AI scraper called FetchFox Hi! I'm Marcell, and I'm working on FetchFox ( https://bit.ly/3z8t1En ). It's a Chrome extension that lets you use AI to scrape any website for any data. I'd love to get your feedback. Here's a quick demo showing how you can use it to scrape leads from an auto dealer directory. What's cool is that it scrapes non-uniform pages, which is quite hard to do with "traditional" scrapers: https://youtu.be/wPbyPSFsqzA A little background: I've written lots and lots of scrapers over the last 10+ years. They're fun to write when they work, but the internet has changed in ways that make them harder to write. One change has been the increasing complexity of web pages due to SPAs and obfuscated CSS/HTML. I started experimenting with using ChatGPT to parse pages, and it's surprisingly effective. It can take the raw text and/or HTML of a page, and answer most scraping requests. And in addition to traditional scraping thigns like pulling out prices, it can extract subjective data, like summarizing the tone of an article. As an example, I used FetchFox to scrape Hacker News comment threads. I asked it for the number of comments, and also for a summary of the topic and tone of the articles. Here are the results: https://bit.ly/3AYS2Ci . You can see the prompt I used for this scrape here: https://bit.ly/3TjL1lZ Right now, the tool does a "two step" scrape. It starts with an initial page, (like LinkedIn) and looks for specific types of links on that page, (like links to software engineer profiles). It does this using an LLM, which receives a list of links from the page, and looks for the relevant ones. Then, it queues up each link for an individual scrape. It directs Chrome to visit the pages, get the text/HTML, and then analyze it using an LLM. There are options for how fast/slow to do the scrape. Some sites (like HN) are friendly, and you can scrape them very fast. For example here's me scraping Amazon with 50 tabs: https://bit.ly/47eVHZ5 . Other sites (like LinkedIn) have strong anti-scraping measures, so it's better to use the "1 foreground tab" option. This is slower, but it gives better results on those sites. The extension is 100% free forever if you use your OpenAI API key. It's also free "for now" with our backend server, but if that gets overloaded or too expensive we'll have to introduce a paid plan. Last thing, you can check out the code at https://bit.ly/3Xaog5i . Contributions welcome :) https://bit.ly/3z8t1En September 4, 2024 at 12:50AM

Show HN: Shelly – A pure and vanilla shell-like interface for the web https://bit.ly/3XtlNUK

Show HN: Shelly – A pure and vanilla shell-like interface for the web shelly is a shell-like inteface for the web made with pure and vanilla HTML, CSS and JavaScript. It's completely configurable and should run decently on any browser. https://bit.ly/4g76acZ September 3, 2024 at 09:43AM

Show HN: Papers from Today https://bit.ly/3MvkKxb

Show HN: Papers from Today https://bit.ly/3XtIqIE September 3, 2024 at 07:08AM

Show HN: Super easy melody loop maker https://bit.ly/3AHdiMP

Show HN: Super easy melody loop maker https://bit.ly/4bZdSTC September 3, 2024 at 04:24AM

Show HN: Passphrase generator that makes an absurd-but-memorable story https://bit.ly/4cYidGU

Show HN: Passphrase generator that makes an absurd-but-memorable story I wanted to share a simple tool I just released: Tool: https://bit.ly/4e84DBo Source: https://bit.ly/3AX9LtP ## Background I have been researching password security and got interested in diceware passphrases ( https://bit.ly/4e5ulXf ). Though the original xkcd comic claims that it is somewhat easy to remember four or five words and turn them into a little scene, I often find it pretty hard to remember the diceware passphrases. (It's taken weeks for me to finally remember it). I found this tool ( https://bit.ly/3AX9wis ) that generates phrases that are more memorable passphrases (original source from Ryan Foster: https://bit.ly/4cSUDLv ). I liked the tool, but I didn't like the interface, so I rewrote the whole thing in React and relaunched it at StrongPhrase.net ( https://bit.ly/3AJTyIq ). ## My goals: * Make the interface a little more intuitive for basic users * Educate folks who are new to security about passphrases, passwords, and password managers. * Learn more myself about entropy, password cracking, and current cracking capacities/times/costs. Example passphrases: * evil juror obtains thin moths * drunk niece and greedy goose clean tall book * emotional boxer and concerned virus acquire 45 smashed baskets I find these a lot easier to remember because they are easier to visualize as a scene! (The downside is that the passphrases are longer to achieve the same level of entropy.) There are many other formats that have varying amounts of entropy. (You can click the "show all 18 formats" in the top right.) I am more of a security hobbyist than an expert. So this project was a chance for me to deepen my learning. Some of my key personal takeaways: * Cloud computing is powerful - I started the project thinking that I'd focus on "time to crack," but a few sources (1Password blog: https://bit.ly/3AX9R4F , Jacob Egner: https://bit.ly/4cOzbY8... ) convinced me that because cloud computing power is so widely available, if you have enough money, you can rent a lot of power and crack much more quickly. * When passphrases are relevant - I already knew passphrases were designed for passwords you actually type. But I found it useful to really get specific about that list: password manager, laptop login password, and perhaps something like Google that you might sign into on new computers from time to time. * Develop my sense of "how much entropy is enough" - The cost to crack table ( https://bit.ly/4cKkR2I ) on the site was helpful in this inquiry. Curious to get feedback/input from this group! Thanks all! https://bit.ly/4e84DBo September 2, 2024 at 11:11PM

Show HN: I built a free tool for breaking down tasks into manageable subtasks https://bit.ly/4dIJONt

Show HN: I built a free tool for breaking down tasks into manageable subtasks Hi HN! First time posting. I created a tool for my Autistic son who is in high school junior battling severe ADHD symptoms. His difficulties include; 1. Time blindness 2. Lack of executive functioning skills 3. Task initiation issues These three were severely affecting his grades. We had a consultation with a Developmental Psychologist who prescribed Concerta that improves his ability to focus in class. In addition he also has the boilerplate ADHD accommodations(2X time on tests and quiet place to take the test). These collectively improved his test taking abilities. But, when it came to his completing his assignments and other submissions "ontime", he is left to his devices. I realized after observing him closely(with his consent :-)), that his real problem was that, the task at hand was overwhelming him. He didnt know how to go about piecing the various sub-tasks that went into completing that task at-hand. For example, for one of his literature class; he had to write a term paper on the book "Rosencrantz and Guildernstern are dead". This was very over-whelming for him which prevented him from starting on-time. I created https://bit.ly/3MsRAif to help him; 1. Manage his workload - Break down his tasks into sub-tasks 2. Plan his work sessions- sign-in using google credentials and lets him load the tasks into his Google calendar 3. Track his progress - Setup alerts(custom alert frequency) to receive alerts on his Phone/PC wherever he is working 4. Stay organized - He can close out the task within the app which then removes the task from calendar and removes the alerts. That gives him a big high knowing he completed a task from his todo-stack. 4. Track time spent- Each subtask has timers that he can start and stop. And there is a dashboard view where he can view metrics on which task he spent the most time on. Here is a snap-shot of one of his working session; https://bit.ly/3Tfdh9n I tried it on one of my own tasks, where I am trying to recreate the Denoising Diffusion Probabilistic Models from scratch and it gave me these steps to follow through; https://bit.ly/3Tfdh9n Tech Stack Flask with JINJA2 templates with Tailwind CSS for styling and SQLite database hosted on Pythonanywhere($5/month hosting and love it) Google Auth for calendar access OpenAI API(GPT4o) for task breakdown Would love for you to try the app and recommend it to others. Would it equally appreciate any bugs. i have tested in on all major browsers, but there might be some lurking bugs somewhere. If you have a specific feature request, i have a repo where I am collecting them. Here is the issues template; https://bit.ly/3Mv08VZ... Cheers, Ananth https://bit.ly/4dIM3Af September 2, 2024 at 07:37AM

Show HN: Ban 'Any updates' emails – this is super-efficient client communication https://bit.ly/4gbu0nK

Show HN: Ban 'Any updates' emails – this is super-efficient client communication Looking for 100 beta users to use RSP for free (in return for great feedback!) I've made a Monday/ Google Sheets hybrid - making it really easy to share structured data with clients, in real-time, with no extra effort from you. This eliminates clients being in the dark or sending multiple emails per task. Project management tools are usually black boxes, where the team only communicates with clients via email threads (in the plural). Instead, with RSP, you choose at the column level what data you want to share with clients and what you want them to edit and not edit, and also have a side-by-side messenger for each task. Client communication is instant, real-time, and super-efficient. https://bit.ly/3Z7axP0 September 2, 2024 at 10:10AM

Show HN: Tech Interview Exp is up and running https://bit.ly/4cV1P9L

Show HN: Tech Interview Exp is up and running https://bit.ly/3PXHc4s September 2, 2024 at 07:58AM

Show HN: Character Headcanon – AI-Powered Backstory Generator https://bit.ly/4cQFXN5

Show HN: Character Headcanon – AI-Powered Backstory Generator Create unique character backstories with Character Headcanon, an AI tool for writers and gamers. Generate instant, detailed profiles for your creative projects. https://bit.ly/3Z3pf9V September 2, 2024 at 12:56AM

Show HN: Server Side Code Integrity https://bit.ly/3XpMqK6

Show HN: Server Side Code Integrity https://bit.ly/4dWYckO September 1, 2024 at 01:48PM

Show HN: I made an app to discover where to stream movies and TV shows in a week https://bit.ly/3Xtj8up

Show HN: I made an app to discover where to stream movies and TV shows in a week I set myself a challenge to make a simple and clear website to discover where to watch movies and TV series within a week. This was my second project with my now favorite tech stack: Next.js, TypeScript, Tailwind and ShadCN UI. https://bit.ly/3MqzM7s September 1, 2024 at 12:29PM

Show HN: I created an extension to give you password requirements on login pages https://bit.ly/3Z6lQXF

Show HN: I created an extension to give you password requirements on login pages Hey, at work, i have to manage everyone's accounts (adobe, microsoft, indeed...) for billing, setting stuff... and it's a nightmare : So many websites with so many different password requirements (minimum characters, lowercase, number...) and none of their login pages give a reminder of the requirements... So, I created Password Requirements Checker, that will give you on the login page requirements for password of the particular website. https://bit.ly/3Mr4JbG August 31, 2024 at 10:17AM

Show HN: PodSnap.AI – automatic AI summaries for your favorite podcasts https://bit.ly/4cIF6hn

Show HN: PodSnap.AI – automatic AI summaries for your favorite podcasts Hi HN, A couple months ago, a friend mentioned that there are too many great podcasts to keep up with. We searched for a solution but didn't find one and realized that many other listeners likely face the same problem. So, I built PodSnap.AI, a service that keeps track of new podcast episodes and automatically sends AI-generated text and audio summaries to your inbox as soon as they are published. This way, users can quickly get the key insights and decide whether to watch the full episode. The service supports podcasts on Apple, Spotify, and YouTube. It's still beta, so I'd highly appreciate any feedback or suggestions for improvement. Thank you! https://bit.ly/3Z2Em3r August 31, 2024 at 10:31AM

Show HN: Supertree – interactive visualization of decision trees in Python https://bit.ly/3yRgk0G

Show HN: Supertree – interactive visualization of decision trees in Python Hi All, I would like to share with you a new Python package for interactive decision tree visualization. It is called `supertree`. It visualizes decision tree as interactive graph, where you can collapse and expand selected nodes. You can zoom and pan though large trees. It works with Scikit-learn, Xgboost, and LightGBM. The package works in notebooks: Jupyter Lab, Jupyter Notebook, Google Colab. You can also use it in Python scripts and save output trees to HTML. The package is available on pip: `pip install supertree`. You can find code examples on the GitHub: https://bit.ly/4e4yY3V Happy exploring! August 27, 2024 at 05:19PM

Show HN: Everything the Bitcoiners Were Promising https://bit.ly/3X7N79x

Show HN: Everything the Bitcoiners Were Promising https://bit.ly/3X56FLV August 29, 2024 at 10:15PM

Show HN: Substack/Medium Writer Personality Test https://bit.ly/3XkWDrs

Show HN: Substack/Medium Writer Personality Test https://bit.ly/3XnXwzx August 29, 2024 at 02:36AM

Show HN: Are You Smarter Than a 5th Grader? https://bit.ly/3ACheOU

Show HN: Are You Smarter Than a 5th Grader? https://bit.ly/3yXtsRM August 29, 2024 at 04:04AM

Show HN: Let me plex that for you – Teach friends and others to use Perplexity https://bit.ly/3AT7GPB

Show HN: Let me plex that for you – Teach friends and others to use Perplexity https://bit.ly/4cG5TKZ August 29, 2024 at 02:55AM

Show HN: IPA, a GUI for exploring inner details of PDF (Rust) https://bit.ly/3YZfucD

Show HN: IPA, a GUI for exploring inner details of PDF (Rust) https://bit.ly/3YZfvgH August 28, 2024 at 11:22AM

Show HN: CMS and SiteBuilder Built with Node.js and AI Based Block Generation https://bit.ly/3MqIbrD

Show HN: CMS and SiteBuilder Built with Node.js and AI Based Block Generation Hi Everyone, I have built a site builder with NodeJS which Uses AI to generate Tailwind based Blocks. You can play with it using https://bit.ly/3MlFnMr link. I didn't enabled Authentication yet. what features do you think is required for a CMS to be usable in real-world projects? What do you think about this project? thanks https://bit.ly/3MpnE6C August 27, 2024 at 08:34PM