10 GITHUB REPOS THAT SHOULD BE ILLEGAL TO HAVE.
all free. all open-source. bookmark this for later.
1. Recordly — the free screen studio.
Open-source screen recorder with auto-zoom, smooth cursor, webcam overlay, styled backgrounds, and polished demos without a video editor. (AGPL-3.0)
https://t.co/aXHOMzGjj3
2. Stirling PDF — your entire PDF toolkit.
Self-hosted. Merge, split, sign, redact, OCR, convert, compress, 50+ tools. Runs locally, so nothing leaves your machine. (MIT)
https://t.co/BIXg2t65Pd
3. PhotoGIMP — turns GIMP into Photoshop.
Photoshop shortcuts, layout, and splash screen patched on top of GIMP 3+.
https://t.co/ZFapnmhNVM
4. Open Notebook — self-hosted NotebookLM.
Drop in PDFs, URLs, and YouTube links. Chat with them, summarize them, and even generate podcasts. Supports 18+ AI providers. (MIT)
https://t.co/aI8ZnHQnVi
5. Odysseus — PewDiePie's self-hosted AI workspace.
Chat, agents, deep research, docs, email, memory, and more. Local-first, your hardware, your data. (MIT)
https://t.co/0OMBoGo5NH
6. FreeDomain — free domain names for everyone.
Register a domain, point it to Cloudflare or any DNS provider, and launch your site without paying for the domain. (AGPL-3.0)
https://t.co/Dki7z6gezx
7. HyperFrames — write HTML, render videos.
HeyGen's open-source engine that converts HTML/CSS animations into deterministic MP4 videos. Built for AI agents. (Apache-2.0)
https://t.co/FgQLeJ9kJ4
8. Web-to-App — turn any website into an Android app.
Configurable WebView, APK signing, and even Node.js, PHP, and Python runtimes. No remote build required.
https://t.co/Ws1TAbl04h
9. Reclip — self-hosted video & audio downloader.
Download videos from YouTube, TikTok, X, Instagram, and 1000+ other sites as MP4 or MP3 using yt-dlp.
https://t.co/XcPKOUuWX1
10. Excalidraw — the infinite whiteboard.
Replace Miro, FigJam, and Lucidchart with hand-drawn diagrams, wireframes, real-time collaboration, and end-to-end encryption. 120k+ GitHub stars. (MIT)
https://t.co/nyi5HkkbWc
Which repo are you trying first?
$411.99 has been auto-distributed to 35 wallets. No claim button. No staking. No lockup.
Here's how @divvyfinance works :
Every trade generates fees → treasury buys a basket of tokenized equities (NVDA, AAPL, MSFT, AMZN, GOOGL + more) → holders get their share automatically, every 2 hours.
Fully onchain.
Fully verifiable.
Next layer : turning x into a divvy interface.
Tag @divvyfinance, run a command → wallet creation, balance checks, mints, contract calls - straight from a tweet.
Some of it's live.
Rest is shipping.
Goal : onchain actions as easy as sending a tweet.
Still early. Mechanism's real. 🛠️
Finch MCP just surprased 430+ User Download this week! last time we reach 15k+ download, can we going higher this szn?
Most AI assistants disappear when the conversation ends. Finch gives them lasting state — memory that accumulates, agents that keep running, vaults that version knowledge, and workflows that continue after you close the chat.
Install : https://t.co/asRgSZARp7
You can now stake FINCH directly from Claude Code
no dashboard, no wallet UI, no app switching, just chat with your agent like you normally would.
Ask for your staking status and get your available balance plus a full breakdown of lock tiers (7d/1x, 30d/1.5x, 90d/2x) in one message, pick the tier that fits your strategy and stake right there in the conversation.
Then turn on auto restake - this is the real upgrade.
Once it's on, your stake silently renews itself at the same tier every time it unlocks.
No gas spent doing it, no manual restaking, no missed cycles because you forgot to check back in time.
Everything runs through the Finch MCP server, live and onchain. Your agent handles the rest, you just ask.
npx -y -p @finchagentic/[email protected] finch install
🚨EL WEB SCRAPING HA MUERTO🚨
Han creado PixelRAG.
Un sistema open source que pasa de parsear HTML.
En vez de convertir una web en texto...
hace una captura de pantalla.
Y luego un modelo vision-language lee la respuesta directamente desde los píxeles.
Brutal.
Porque el scraping tradicional rompe muchísima información:
→ tablas
→ gráficos
→ layouts
→ columnas
→ botones
→ PDFs
→ páginas mal estructuradas
Todo eso se pierde cuando conviertes una web en texto plano.
PixelRAG hace algo mucho más simple:
indexa la página tal y como la ve una persona.
El equipo ha creado un índice visual de Wikipedia entero:
+30 millones de screenshots.
Y aun así supera al mejor RAG basado en texto por un 18,1% en QA.
Tiene también un plugin para Claude Code.
Básicamente le da ojos a Claude.
Le pasas una URL, un paper de arXiv o tu web local…
y Claude no scrapea el DOM.
La mira.
Cómo funciona:
→ renderiza webs, PDFs e imágenes en tiles
→ los embebe con Qwen3-VL-Embedding
→ construye un índice FAISS
→ busca visualmente sobre píxeles
Todo open source bajo Apache-2.0.
Link al repo en los comentarios 👇🏻
We're integrating with @RobinhoodApp new AI agent stack.
excited to bring NoelClaw's runtime, memory, and automation layer into that ecosystem.
Time to build.
Anthropic just dropped 5 workshops on building self-improving agentic systems from scratch:
00:00 - Ship your first Claude agent
36:44 - Build memory for Claude agents
1:05:06 - Make your agent autonomous
1:26:46 - Set up a proactive agent
2:03:35 - self-improving agents (tools,skills)
These 3-hours of free Claude workshops will replace 10 paid agentic courses.
Watch today, then read article below on how to build a self-improving agentic system with Fable 5.