Joined January 2012
Someone wrote a headless browser that renders pages with zero Chromium in the stack. It's 70MB. Chrome is 300MB+. It's called Obscura, a headless browser rebuilt from scratch in Rust. And it runs your existing Puppeteer and Playwright scripts unchanged. No Chromium. No Node.js. No dependencies. One binary. ↓ The numbers that matter: Memory: 30MB vs Chrome's 200MB+ Binary size: 70MB vs 300MB+ Page load: 84ms on a JS + XHR + fetch page vs Chrome's ~800ms Startup: instant vs ~2s It runs real JavaScript through V8 and speaks the full Chrome DevTools Protocol, so it's a drop-in replacement. Point your Puppeteer connect() at ws://127.0.0.1:9222 and everything just works. The part that's actually wild: native rendering with no Chromium underneath. Screenshots, full-page captures, live screencasting, and PDF export all run on Obscura's own rendering engine written from scratch. Block, flex, grid, tables, transforms, SVG, canvas. No Chrome process anywhere in the stack. Then there's stealth mode, which is where the scraping crowd will lose it: Per-session fingerprint randomization (GPU, screen, canvas, audio, battery) navigator.webdriver set to undefined, matching real Chrome event.isTrusted = true on dispatched events Native function masking so toString() returns [native code] 3,520 tracker domains blocked automatically It also ships an MCP server, so you can wire it straight into Claude Desktop or Cursor and let an agent navigate, click, fill forms, screenshot, and pull page text as browser tools. Install is one curl command. Docker image is ~57MB compressed on distroless, no shell, no package manager. Apache 2.0. 22.1k stars. The maintainer says the open-source engine stays fully featured with no feature gating, ever. If you're running scraping jobs or agent automation on headless Chrome right now, this is worth a weekend. github.com/h4ckf0r0day/obscu…
37
79
5
884
65,011
The winner of an Anthropic hackathon open sourced his entire Claude Code setup - and this is absolute f*cking gold 68 subagents, 286 skills, 94 commands, MIT license ECC swaps a lone Claude Code assistant for a whole engineering department a blueprint comes before any build, a failing test comes before any fix, and every change gets a second look from a context that never saw it written • who does what > planning - hand it a single sentence, get back a plan you sign off before a line of code > review - a fresh-context read of your diff, with its own reviewer for each language > build repair - a dedicated fixer per toolchain, PyTorch and CUDA included > security - an OWASP sweep, plus a scanner hunting injection holes in your agent config > architecture - catches design mistakes while they're still cheap, before they turn into migrations > domain work - database queries, ML pipelines, e2e tests, docs the security duo is the piece most people never set up an external OWASP audit costs four figures and a week of waiting this one wraps up on your branch by lunchtime • what the skills cover > testing - tdd-workflow moves you from red to green, eval-harness runs above it > language packs - Python, Go, Rust, C++, Django, Laravel, Spring Boot, Next.js > context - search-first checks the docs before writing, iterative-retrieval stops your whole repo flooding the window > shipping - Docker, CI/CD, health checks, rollbacks, migrations > beyond code - writing in your voice, market research, pitch decks fork it, trim it, and have your own version running by tomorrow begin with a single plan and a single rules pack turning on all 286 skills at once is the quickest way to make it worse whoever wires this in over a weekend spends the next quarter reviewing work instead of typing it
82
319
17
2,453
373,074
10个GitHub仓库,免费到离谱! 先收藏。 1.Recordly 开源录屏。自动变焦、鼠标跟得住、摄像头叠上去,演示不用再丢给剪辑。AGPL。 github.com/webadderallorg/Re… 2.Stirling-PDF 自己架的PDF工具箱。合并、拆分、签名、涂黑、识别文字、压缩,五十多种。文件不出这台机器。MIT。 github.com/Stirling-Tools/St… 3.PhotoGIMP 给GIMP打补丁,快捷键和布局往Photoshop靠。从PS过来的人少踩一会儿坑。 github.com/Diolinux/PhotoGIM… 4. Open Notebook 自己架的NotebookLM。丢PDF、网页、YouTube链接,能对话、能摘要、能出播客。模型自己选,十几家都能接。MIT。 github.com/lfnovo/open-noteb… 5.Odysseus PewDiePie那套自托管AI工作区。聊天、Agent、调研、文档、邮件、记忆,数据留在自己硬件上。AGPL。 github.com/odysseus-dev/odys… 6.FreeDomain 免费域名项目。注册后指到Cloudflare或自己的DNS就能挂站。AGPL。给的是他们平台上的免费域,不是任意.com。 github.com/DigitalPlatDev/Fr… 7.Hyperframes HeyGen开源的。HTML和CSS写成片子,同样的代码吐出同样的MP4。给Agent用的。Apache-2.0。 github.com/heygen-com/hyperf… 8.Web-to-App 网页在本机打成安卓应用。可配WebView、签名,还能带上运行时。原帖是短链,发之前点开核对一次。 9.Reclip 自己架的下载页。粘YouTube、TikTok、X、Instagram链接,出MP4或MP3。底层是yt-dlp。只适合自己存档,站点条款自己看。 github.com/averygan/reclip 10.Excalidraw 手绘风白板。对Miro、FigJam、Lucidchart那一类。实时协作,端到端加密。MIT。 github.com/excalidraw/excali… 很多人按月付钱的功能,Github里已经有了。 开源换的是自己说了算,界面未必有付费软件那么靓。 会动手的人,这清单够用了!
8
102
1
645
40,230
🚨 Google acaba de liberar una de sus herramientas internas más potentes. Ha lanzado Magika, un sistema que detecta el contenido real de cualquier archivo, aunque esté disfrazado. Por ejemplo, malware que parece un PDF o scripts ocultos como imágenes. Es gratis y open-source 👇
44
364
18
2,732
188,788
Claude built your app in 2 hours. Give it 20 more minutes so it doesn't get hacked in 2 days. Ask Claude to: - Hide API keys - Purge secrets from Git - Expose only the public DB key - Enable row-level security - Encrypt sensitive data - Enforce server-side auth - Lock record access - Block field tampering - Secure session cookies - Hash passwords - Rate limit login - Add bot protection - Parameterize queries - Validate all input - Escape user content - Restrict file uploads - Trim API responses - Add security headers - Force HTTPS - Scan dependencies Vibe coding is fun. Getting hacked isn't 🔒
19
146
1,757
153,662
Liberta retweeted
You can now scrape the web with AI agents in 85ms. Someone built in Rust what Chrome engineers couldn't optimize in years. Obscura is an open-source headless browser engine for web scraping and AI automation. A 70MB binary that boots instantly, loads pages in 85ms, and runs on 30MB of RAM. The Google equivalent needs 300MB of disk, 200MB of memory, and 2 seconds to start. The real trick is stealth mode. Per-session fingerprint randomization across: > GPU, canvas, audio, battery > 3,520 tracker domains blocked > navigator.webdriver fully masked > Native functions patched cleanly No detector can catch it. The engine runs real JavaScript through V8 and speaks the Chrome DevTools Protocol natively. That means it works as a drop-in for Puppeteer and Playwright. You swap the binary, your code stays untouched. One developer shipped this alone. Free to use.
49
257
7
2,559
167,024
Liberta retweeted
Stop renting a chatbot and calling it your company’s brain. Utopia is the first open-source enterprise world model I’ve seen that actually treats knowledge like a system, not a vibe. It's a Self-hosted, open source and built around the one question every other knowledge base ignores: when. Every fact in it carries a start date and an end date. Nothing gets overwritten. When something changes, the correction closes the old version and opens a new one. The history stays intact. So the graph isn't a snapshot anymore. It's a recording you can rewind. You drag a timeline on the screen and watch the whole thing redraw itself to show what was true on that date. Who owned what. What the policy said before it changed. All of it, replayable. And every fact points back to the exact sentence it came from. Nothing gets in without a receipt. Here's the part that got me. The whole thing is one binary and one Postgres database. That's it. No Elasticsearch, no separate vector service and no message queue. Most RAG stacks are six services duct-taped together. This is two. It even runs fully offline on an air-gapped network with any model you want. github.com/deeplethe/utopia
79
299
12
2,420
176,400
Liberta retweeted
The winner of an Anthropic hackathon open sourced his entire Claude Code setup. 68 subagents, 286 skills, 94 commands, MIT license ECC turns Claude Code from one assistant into a full engineering team. It plans before it builds, writes the failing test first, then reviews its own work from a fresh context. The agents, by job: Planning. Turns a one-line request into a blueprint you approve before any code exists. Review. Reads your diff in a clean context, with separate reviewers for Go, Python, TypeScript, Rust, Java and more. Build repair. One resolver per toolchain, down to PyTorch and CUDA errors. Security. An OWASP pass on your code, plus a scanner that audits your own agent config for injection risks. Architecture. System design calls before they turn into migrations. Domain work. Database queries, ML pipelines, end-to-end tests, docs. The skills, by category: Testing. tdd-workflow gates you from red to green, with eval-harness and verification-loop sitting on top of it. Language packs. Idioms, testing and security for Python, Go, Rust, C++, Django, Laravel, Spring Boot, Next.js. Context. search-first makes it read the docs before writing, iterative-retrieval keeps subagents from dragging your whole repo into the window. Shipping. Docker, CI/CD, health checks, rollbacks, and migration patterns for Prisma, Drizzle and Django. Work that is not code. Writing in your voice, market research with sources, pitch decks, slide decks. Start with one plan and one rules pack. Installing all 286 skills at once is the fastest way to make it worse.
93
313
13
2,758
414,515
I still don't understand why everyone is not using this yet. Thanks to it, a year ago I increased my income to 17,000 dollars a month Andrey Karpathy, co-founder of OpenAI, published a simple idea that got 16 million views: stop using AI to write code, use it to build a second brain You point Claude Code to a folder, drop any source in there (an article, transcript, PDF) and Claude reads it, links it, and saves it into a living wiki of everything you know. It compounds like interest: the more you feed it, the smarter it gets Here is the gist: Install Obsidian, create a vault, open it in Claude Code Paste the file with Karpathy's wiki idea and tell Claude to build it Claude creates three folders: raw for sources, wiki for its pages, CLAUDE which runs everything Drop any source into raw and say "ingest this" Ask questions across everything, forever Five minutes to set up, and you never start with an empty chat again The full step-by-step guide is in the article. Save to bookmarks
367
1,477
53
11,395
2,883,057
DeepSeek just killed the coding-agent industry. deepseek-harness: Open-source. Free. Claude Code's top plan costs $200/month. What it does: → Everything is a plugin - model adapters, tools, session logs, even the agent loop itself, all swappable from config → Built on Cordis, DeepSeek's own plugin runtime → One command to run: npx @deepseek-ai/dsh web - spins up a local Web UI → Ships with real architecture docs + an AGENTS.md written for agents to read the codebase themselves → TypeScript, MIT, developer preview - breaking changes expected, DeepSeek says so themselves What it replaces: → Anthropic's Claude Code → Closed, single-vendor coding-agent environments generally A frontier lab just gave away, for free, the exact layer companies charge a subscription for. → github.com/deepseek-ai/deeps…
83
324
17
2,402
293,502
Малоизвестное интервью Михаила Горбачёва, в котором он прошёлся по путину и дал тому полную характеристику – закончит катастрофой. Собственно, так оно и вышло. Что интересно, очень многие «патриоты» ненавидят Горбачёва, за то что тот дал им свободу, позволил заработать денег и открыл границы, и обожают путина, за то что тот превратил страну в международного изгоя и начал войну.
123
848
25
3,868
98,812
Liberta retweeted
I spent two years taking notes in Obsidian.. ..and treating Claude like a search engine and the frustration of that combination is that both were powerful and completely useless to each other at the same time. Four steps and fifteen minutes connected them permanently with no coding and no course required. My article covers the full method I built on top of that foundation for a second brain that works 24 hours a day. Full guide below.
11
51
3
230
42,940
I found 10 GitHub repos that can scrape literally the entire internet for free. Companies charge $2,000/month for API access to do exactly this. It's all open source. Bookmark this thread. 🧵👇
3
22
67
6,798
Liberta retweeted
THE CREATOR OF OBSIDIAN JUST SHIPPED HIS OWN CLAUDE SKILLS PACK 44,000 developers already installed it. most people building second brains with claude have no idea it exists these aren't community skills. they're written by kepano himself - the person who built obsidian. that means they respect obsidian's actual conventions: markdown, bases, JSON canvas, backlinks, plugins, all of it drop them into your claude skills folder once. next time you say "summarize my notes about karpathy" or "extract this article to my inbox," the right skill wakes up automatically. no prompt engineering. no manual context. no config i tested 10 claude + obsidian repos to write my full breakdown. this is the one with the highest trust factor - because the person shipping it is the one who defined what an obsidian vault even looks like the full ranked breakdown of all 10 is in the article i test things like this on myself and share the results. follow @0xkkai if you want to see what's next
39
150
3
1,196
257,379
10个GitHub仓库,能爬遍整个互联网 全部收藏。每个都能把任何网站的数据扒得干干净净。这种权限通常得走销售渠道、签合同才能有。
19
585
9
3,222
201,124
50个看似"违法"但完全合法的网站 1. cobalt.tools 下载任何社交媒体视频 2. photopea.com 免费Photoshop 3. temp-mail.org 一键临时邮箱 4. tinywow.com 100+免费工具合集 5. archive.org 访问任何旧网页 6. libgen.li 数百万免费教科书 7. sci-hub.al 免费研究论文 8. alternativeto.net 查找免费应用替代品 9. justwatch.com 定位任何内容的流媒体位置 10. gutenberg.org 7万本免费经典书籍 11. pdfdrive.com 免费PDF下载 12. openculture.com 精英大学免费课程 13. wolframalpha.com 瞬间解决任何数学问题 14. remove.bg 一键去除背景 15. cleanup.pictures 从照片中删除物体 16. unscreen.com 免费视频去背景 17. squoosh.app 免费压缩任何图像 18. excalidraw.com 免费手绘图表 19. carbon.now.sh 将代码转换为艺术作品 20. ray.so 令人印象深刻的代码截图 21. flightradar24.com 实时追踪任何航班 22. camelcamelcamel.com 追踪亚马逊价格历史 23. haveibeenpwned.com 检查是否被黑客入侵 24. virustotal.com 扫描任何文件的恶意软件 25. privnote.com 发送自毁消息 26. file.io 分享自动删除的文件 27. archive.ph 永久归档任何网页 28. accountkiller.com 从任何网站删除你的信息 29. radio.garden 收听全球任何电台 30. tunefind.com 识别任何节目的歌曲 31. musicforprogramming.net 专注音乐 32. mynoise.net 个性化专注音景 33. annasarchive.org 搜索每本已出版的书 34. elicit.com 研究论文AI助手 35. consensus.app 搜索科学共识 36. connectedpapers.com 可视化研究地图 37. semanticscholar.org 免费学术搜索 38. scispace.com 理解任何研究论文 39. summarize.tech 总结YouTube视频 40. phind.com 开发者AI搜索 41. regex101.com 即时测试正则表达式 42. codebeautify.org 清晰格式化任何代码 43. explainshell.com 理解终端命令 44. tldraw.com 浏览器无限画板 45. downdetector.com 检查网站是否宕机 46. tineye.com 反向图像搜索 47. fast.com 测试网速 48. smallpdf.com 免费编辑PDF 49. ilovepdf.com 合并和分割PDF 50. 10minutemail.com 数秒内获得临时邮箱 全部合法。全部免费。 收藏以免遗忘。🔖
136
1,752
53
9,006
788,199
Liberta retweeted
Directorio de cosas útiles (actualizado) duckduckgo.com buscador que respeta privacidad massgrave.dev activar windows, ISOs, activar office sci-hub.box leer cualquier paper académico gratis cobalt.tools descargar videos de twitter e instagram silent.link eSIM internacional anónima fitgirl-repacks.site juegos crackeados simplelogin.io aliases de e-mail ilovepdf.com convertir, cortar, juntar, todo con PDFs opensubtitles.org subtítulos en todos los idiomas subdivx.com subtítulos en español y argentinos wayback.archive.org páginas web archivadas archive.is sacar el paywall de artículos incompetech.com música sin derechos de autor freesound.org SFX sin derechos de autor android.com/find encontrar tu teléfono perdido linguee.com traducciones y ejemplos roadmap.sh roadmaps para aprender informática erowid.org información sobre drogas argenpills.org información sobre drogas argentina retroflix.org ver películas del dominio público nic.ar registrar dominios .com.ar o .ar spaceship.com registrar dominios cualquier TLD photopea.com alternativa online a photoshop tineye.com buscar una imagen en distintas resoluciones openlibrary.org libros gratis sin derechos annas-archive.pk libros gratis con y sin derechos rufus.ie formatear y flashear pendrives wordcounter.net contador de palabras calculator.net calculadora fast.com chequear velocidad de tu internet openspeedtest.com lo mismo pero sin flash ni java convertio.co convertir archivos de un formato a otro libreoffice.org alternativa libre y gratis a MS Office qr.grid.ws generador de QR bit.ly acortador de links handbrake.fr convertir formatos y códecs de video tinypng.com compresor de imágenes userbenchmark.com comparar builds de PC nextcloud.com almacenamiento en la nube open source wiki.archlinux.org la wiki de arch linux
39
679
17
5,649
212,127
Liberta retweeted
Ex-Google engineer just released a free 3-hour course on building and monetizing AI agents. How to go from one agent to a full system that finds leads and makes money: 00:00 - Design an AI agent system 07:38 - Add human handoffs 19:27 - Understand RAG and vector databases 52:38 - Deploy agents to Google Cloud 1:25:57 - Turn agents into a paid WhatsApp business 1:33:00 - Convert conversations into leads 2:24:01 - Learn loops vs graphs 2:33:08 - Build a multi-tool agent graph Most people are still building AI demos. This course shows the full stack: Agents → RAG → Deployment → Leads → Revenue Building agents is the old workflow. Monetizing agent systems is the new one. This 3-hour watch covers more than most $500 paid courses. Save it before everyone starts selling the same thing.
50
770
14
4,021
354,618
Liberta retweeted
EL MODELO MÁS GRANDE DEL MUNDO OPEN SOURCE CORRE EN MENOS DE 4GB DE VRAM 2.8 billones de parámetros. GPU de 4GB. Sin cuantización. Sin destilación. Sin pruning. Cómo lo hace: Solo carga una capa a la vez en la GPU. En los modelos MoE (como Kimi K3) solo carga los expertos que el token necesita. Resultados: • Llama 70B → 4GB • Llama 405B → 8GB • DeepSeek-V3 (671B) → 12GB • Kimi K3 (2.8T) → <4GB Se llama AirLLM. Open source. 24.5k estrellas. La mayoría sigue creyendo que necesita una H100. pero ya no. Repoo 👇
43
261
10
1,847
120,805