GitHub - ARahim3/DigUp: Search your computer by what's in your files: photos, PDFs, documents, recordings, videos and code. On-device on Mac, Windows and Linux, built on Google DeepMind's EmbeddingGemma 2. · GitHub ARahim3/DigUp
Describe the file you lost and DigUp finds it, fully on-device, down to the exact PDF page or video moment. 🔗 ARahim3/DigUp
Credits: ARahim3/DigUp
#digup #swift #semanticsearch
These Free GitHub Repos Feel Illegal to Use (Clone Viral Videos, Replace Mailchimp)
Five free apps that feel like they shouldn't be allowed. One clones any viral video's whole format in one command. One cuts your newsletter bill down to a few dollars of email. One lets you search your own computer like Google, by what's inside your files, without anything leaving your laptop. I installed all five on my own Mac and tested every one on demo files. Here's what works, how long setup really takes, and the catch on each.
📄 FREE: The Loophole List: all five download links, the exact commands I ran, setup times, my test results and every catch, in one PDF. Comment LOOPHOLE and I'll send you the free Loophole List.
▶️ Tip: watch at 1.25× or 1.5× speed for the best experience.
⏱ Chapters
0:00 Five free apps that feel illegal
0:44 How I rate them: works, half works, hype
1:01 #5 Filee: convert any file, offline
2:06 #4 DigUp: search your files by what's in them
3:21 #3 ai-memory: one memory for every AI tool
4:37 #2 MailCheap: the newsletter app that started a fight
6:08 #1 Hypit: clone any viral video
7:26 Copying a format: fair or not?
8:00 The scoreboard
8:35 Your turn
🧰 The 5 free repos
• #5 Filee, a drag-and-drop file converter that works offline: KnifeLemon/Filee
• #4 DigUp, search your files by what's in them, on your own laptop: ARahim3/DigUp
• #3 ai-memory, one shared memory for Claude Code, Codex and 20+ AI tools: akitaonrails/ai-memory
• #2 MailCheap, a newsletter app where you pay per email sent: marclou/mailcheap
• #1 Hypit, clone any viral video's format with your AI coding agent: hypit-ai/hypit
📊 My results (one Apple Silicon Mac, Oct 25, 2026)
• Filee: WORKS. Under 1 min to set up. iPhone photo to JPG 0.51 s, Word to PDF 0.83 s, 3 PDFs merged 0.26 s, with the internet blocked.
• DigUp: WORKS. 3 min to set up (865 MB model download). 3 of 3 plain-English searches right on the first try.
• ai-memory: WORKS. About 2 min to set up. Codex picked up a Claude Code session, including the failed fix and the missing cent, without me re-explaining.
• MailCheap: HALF WORKS. 3 min to the dashboard, Mailchimp import worked, but nothing sends until Amazon SES approves you, and self-hosting has no support.
• Hypit: WORKS. A finished clone in 10 min for $0 with code-drawn visuals. AI people or footage need its paid service (about $1.15 a video, plans from $9.90/mo).
🎬 Don't want to touch a coding agent? We built Make Viral Reels: paste any link (or pick a story going viral in your niche), choose a style, and get a finished, ready-to-post reel with voice, captions and music. No face, no editing. Your first video is free:
makeviralreels.com
Honest notes:
• Pricing: Mailchimp at 5,001 to 10,000 contacts starts at $110/mo (Essentials, regular price); beehiiv at 10,000 is $119/mo (Lite, yearly). Both checked logged out on Oct 25, 2026.
• Credits: the Filee and DigUp app demo footage is from each project's own README (KnifeLemon/Filee and ARahim3/DigUp, MIT). The tier-list reference clip is Hypit's own example from its README. Tweets shown are public posts by Marc Lou and Tyler Denk.
• Every test used made-up demo files in a test folder, never real personal files. One test per project, not a study.
📚 Go deeper with my guides:
• The Complete Claude Code Guide: hyperautomationlabs.gumroad.com
• The OpenAI Codex Guide: hyperautomationlabs.gumroad.com
• Claude Cowork Sales Guide: hyperautomationlabs.gumroad.com
• Claude Certified Architect Prep Kit: hyperautomationlabs.gumroad.com
❤️ Support the channel: hit Join on YouTube to become a member, or become a Supporter on Facebook.
Deep dives like this live here on YouTube → youtube.com
Bite-size AI drops on Instagram → instagram.com
Join the community on Facebook → facebook.com
Nothing in this video is sponsored. None of these projects paid to be included, and I'm not affiliated with any of them. Make Viral Reels is my own product.
#GitHub #OpenSource #FreeApps #ClaudeCode #AI
216 likes · 41 comments · 14,994 views
Ich habe 50+ Github Repos getestet. Das sind die 6 BESTEN (Oktober 2026)
🚀 AUTelligence Solutions (KI-Lösungen für Unternehmen):
autelligence-solutions.com
📞 Kostenloser Call mit mir (arbeite mit mir zusammen):
cal.flowwy.de
🎓 Meine Community (alle Ressourcen & Templates):
skool.com
🔗 Alle GitHub Repos aus dem Video:
▪️ Claude Motion Kit (mein eigenes Repo, einzigartige Motion Graphics mit Claude):
nikisge/claude-motion-kit
▪️ Insomnia (Agenten arbeiten bei zugeklapptem MacBook weiter):
krishhgg/Insomnia
▪️ Security Audit Skill von Cloudflare (ein Team aus KI-Agenten prüft deinen Code auf Sicherheitslücken):
cloudflare/security-audit-skill
▪️ Strata (Qwen 3.8 Flash Next lokal auf deinem Gaming-PC):
Niko1221/Strata
▪️ DigUp (deine Dateien nach Inhalt durchsuchen, komplett lokal):
ARahim3/DigUp
▪️ God's Eye View (die ganze Welt live auf einem 3D-Globus):
bilawalsidhu/gods-eye-view
🛠️ Extras aus dem Video:
▪️ HyperFrames (macht aus HTML ein Video, Basis für das Motion Kit):
heygen-com/hyperframes
▪️ What Ships (Launch-Videos als Referenzen für Claude):
whatships.com
In diesem Video zeige ich dir 6 GitHub Repos, die meine Arbeit mit KI komplett verändert haben – und alle kannst du kostenlos nutzen.
Egal ob du mit Claude coole Videos und Motion Graphics erstellen willst, dich für lokale KI interessierst oder viel mit Claude Code und Codex arbeitest: Hier ist für dich etwas dabei.
Du siehst:
▪️ wie du mit meinem eigenen Claude Motion Kit und HyperFrames Motion Graphics baust, die nicht nach KI-Slop aussehen (mit echten Referenzvideos, Storyboard und Review-Phase)
▪️ wie Insomnia deine Coding-Agenten weiterarbeiten lässt, während dein MacBook zugeklappt ist
▪️ warum gerade so viele Sicherheitslücken auf einmal auftauchen (PHP, Python, Node.js) und wie der Security Audit Skill von Cloudflare deinen Code von einem ganzen Team aus KI-Agenten prüfen lässt
▪️ wie Strata das 125B-Modell Qwen 3.8 Flash Next schon ab 12 GB Grafikspeicher und 32 GB Arbeitsspeicher lokal laufen lässt – und warum Mixture of Experts das so schnell macht
▪️ wie du mit DigUp deine Videos, Audios, Bilder und PDFs nach Inhalt durchsuchst, komplett lokal auf deinem Mac
▪️ wie du mit God's Eye View Flüge, Verkehr und Live-Kameras auf der ganzen Welt verfolgst – sogar per Sprachsteuerung
📹 Strata im ausführlichen Test bei Fabian:
Lokale KI: 125B-Modell auf einer Gaming-Grafikkarte? Strata im Test
youtu.be
⏱️ Zeitstempel:
0:00 Einstieg
0:37 1. GitHub Repo
5:04 2. GitHub Repo
6:18 3. GitHub Repo
9:06 4. GitHub Repo
11:39 5. GitHub Repo
13:13 6. GitHub Repo
16:01 Abschluss
💬 Welches der 6 Repos probierst du als Erstes aus?
Und wenn du ein ausführliches Video zum Claude Motion Kit willst, schreib es mir in die Kommentare.
#ki #github #opensource #claudecode #codex #motiongraphics #hyperframes #lokaleki #kiagenten #cybersecurity #ai #tools #kostenlos
[en-US] I Tested 50+ GitHub Repos. These Are the 6 BEST (October 2026)
🚀 AUTelligence Solutions (AI solutions for businesses):
autelligence-solutions.com
📞 Free call with me (work with me):
cal.flowwy.de
🎓 My community (all resources & templates):
skool.com
🔗 All GitHub repositories from the video:
▪️ Claude Motion Kit (my own repository, unique motion graphics with Claude):
nikisge/claude-motion-kit
▪️ Insomnia (agents work with a closed MacBook) (continued):
krishhgg/Insomnia
▪️ Cloudflare Security Audit Skill (a team of AI agents checks your code for vulnerabilities):
cloudflare/security-audit-skill
▪️ Strata (Qwen 3.8 Flash Next locally on your gaming PC):
Niko1221/Strata
▪️ DigUp (search your files for content, completely locally):
ARahim3/DigUp
▪️ God's Eye View (the whole world live on a 3D globe):
bilawalsidhu/gods-eye-view
🛠️ Extras from the video:
▪️ HyperFrames (turns HTML into video, basis for motion) Kit):
heygen-com/hyperframes
▪️ What Ships (launch videos as references for Claude):
whatships.com
In this video, I'll show you six GitHub repositories that have completely transformed my work with AI—and you can use them all for free.
Whether you want to create cool videos and motion graphics with Claude, are interested in local AI, or work extensively with Claude Code and Codex: there's something for you here.
You'll see:
▪️ how to build motion graphics that don't look like AI slop using my own Claude Motion Kit and HyperFrames (with real reference videos, storyboard, and review phase)
▪️ how Insomnia lets your coding agents keep working while your MacBook is closed
▪️ why so many security vulnerabilities are suddenly appearing (PHP, Python, Node.js) and how Cloudflare's Security Audit Skill lets an entire team of AI agents check your code
▪️ how Strata makes the Qwen 3.8 Flash Next 125B model run locally with as little as 12 GB of graphics memory and 32 GB of RAM – and why Mixture of Experts makes it so fast
▪️ how to search your videos, audio, images, and PDFs by content using DigUp, all locally on your Mac
▪️ how to use God's Eye View to view flights, traffic, and live cameras Tracked worldwide – even via voice control
📹 Strata in an in-depth review by Fabian:
Local AI: 125B model on a gaming graphics card? Strata put to the test
youtu.be
⏱️ Timestamps:
0:00 Introduction
0:37 1st GitHub Repo
5:04 2nd GitHub Repo
6:18 3rd GitHub Repo
9:06 4th GitHub Repo
11:39 5th GitHub Repo
13:13 6th GitHub Repo
16:01 Conclusion
💬 Which of the 6 repos will you try first?
And if you'd like a detailed video about the Claude Motion Kit, let me know in the comments.
#AI #GitHub #OpenSource #ClaudeCode #Codex #MotionGraphics #Hyperframes #LocalAI #AIAgents #Cybersecurity #AI #Tools #Free
132 likes · 14 comments · 5,834 views
DigUp Searches Local Media, Documents and Code On-Device
DigUp is available as a free, open-source desktop application for searching selected local documents, images, recordings, videos and source code. Its macOS build is accompanied by beta releases for Windows and Linux. For developers, the immediate consequence is one search interface that can locate a function, screenshot, PDF page or point in a recording without uploading those files to a hosted service, according to the project repository.
One query, two retrieval paths
DigUp runs Google DeepMind’s EmbeddingGemma 2 through llama.cpp. Its readers convert files into text, pictures or sound representations, then store vectors for pages, passages, images, video frames and audio segments in a SQLite index. Those vectors occupy the same 768-dimensional space, allowing one text query to rank different media types together, according to the implementation description.
Semantic retrieval is only one part of the search path. DigUp also maintains a keyword index and boosts exact-word matches, which is intended to preserve literal searches for names, invoice numbers and error codes that an embedding may not represent precisely. Queries therefore combine vector proximity with lexical matching rather than treating embeddings as a substitute for exact search, as the README explains.
Indexing granularity depends on the format. DigUp processes PDFs page by page, splits audio into overlapping 30-second windows, and samples video frames while also indexing the soundtrack. Search results can point to a PDF page or media timestamp, although jumping directly to that location depends on the viewer available on each operating system, according to the supported-format and playback notes.
Code stays in a separate index
Code search is disabled until the user enables it for selected folders. It reads source files, notebooks without their outputs and repository documentation, then opens matches at the relevant line in a configured editor. The code index remains separate from general file search, and code queries must begin with code:, according to the code-search documentation.
The project says this path skips files ignored by Git, along with vendored or generated code, data files and content that resembles a key or password. It lists support for editors including VS Code, Cursor, Zed, Sublime Text and JetBrains IDEs, plus several platform-specific options, in the same documentation.
Local processing still has a cost
The maintainers state that indexed content does not leave the computer and that DigUp has no account requirement or telemetry. The application connects to download its model once and to check for updates once per day; the update check can be disabled and is described as sending no information about indexed files. These are claims made by the project itself.
That design shifts computation and storage onto the user’s machine. The initial model download is 865 MB, and the first indexing pass can take from minutes to hours depending on the selected files and hardware. DigUp estimates that duration during setup; on Windows and Linux it uses a GPU when a setup check finds that path faster than the processor while producing the same results, according to the installation and performance notes.
Platform constraints are uneven. The Mac version requires Apple Silicon and macOS 14 or later. Windows requires 64-bit Windows 10 or 11, and its beta installer is not yet signed. The documented Linux packages target 64-bit Intel and AMD systems, with .deb and AppImage distributions available from the project’s releases and installation instructions.
Analysis: broader retrieval means broader access
Analysis: DigUp’s consequential design choice is the combination of cross-media vector ranking with exact-string retrieval, not semantic search alone. That can make fragmented project material searchable through one interface, while the separate code index limits accidental mixing between repositories and personal files. The architecture and separation are documented in the repository.
The unresolved trade-off is scope. Better retrieval requires granting the application access to more folders and paying the corresponding indexing cost. Folder selection, opt-in code search and exclusions for several sensitive file categories narrow that boundary, but they do not remove the need to decide which local material belongs in a consolidated search index. Those controls and exclusions are described by the project; judging whether they are sufficient remains a user-side trust decision.
DigUp: Describe it. Dig it up. Search your computer by what's in your files (4.6k stars)
What it does
All text below is quoted directly from the project README.
>Search your computer by what's in your files: photos, PDFs, documents, recordings, videos and code. On-device on Mac, Windows and Linux, built on Google DeepMind's EmbeddingGemma 2.
What you can find:
- "zebra in a video" opens the clip at the moment the zebra shows up
- "where they talk about sleep" jumps to that minute of a podcast
- "the clause about pets in the lease" shows the PDF page
- "a dog on the beach" finds the photo, and "payment declined error" the screenshot
- a search in English can find a note written in Bengali, Arabic or any of the model's 100+ languages, and the other way around
- "code: retry a failed call with backoff" finds the function in your projects, if you turn on code search
>Nothing you index leaves your computer. DigUp goes online only to download the model once and to check for updates once a day. The update check sends nothing about your files, and you can turn it off. No account, no telemetry.
>Mac. Download the DMG from Releases and drag DigUp to Applications, or use Homebrew: brew install --cask arahim3/tap/digup. It needs an Apple Silicon Mac with macOS 14 or later.
Repository
ARahim3/DigUp
Repository stats
|Metric|Value|
|:-|:-|
|Stars|4,631|
|Forks|333|
|Last release|v0.6.1 (2026-10-12)|
|Languages|Swift 66.5%, Rust 12.3%, JavaScript 8.7%|
|Contributors|4|
|AI contributor accounts (Claude / Codex / Cursor / Copilot)|None|
|AI co-authored commits|None|
Main contributor
|Field|Value|
|:-|:-|
|Name|Abdur Rahim|
|GitHub|@ARahim3|
|Commits to this repo|142 (top contributor)|
|X / Twitter|@ARahim|
|Followers|233|
12 points
No telemetry.
It's free and open source.
🔗 ARahim3/DigUp
Imagine never hunting through folders again.
#AI #OpenSource #Productivity #DeveloperTools
1 like
Repos de GitHub para la IA: top 3 de esta semana
Tres repos de GitHub de esta semana si usas la IA: una skill que te explica cosas construyéndolas, un buscador que encuentra el segundo exacto de un vídeo y una placa de 7 $ con la que una IA usa tu iPhone.
Anatomy: wheresryan22/anatomy
DigUp: ARahim3/DigUp
ESP32 iPhone Agent: onehumanbeing/esp32-iphone-agent
#github #ia #claudecode #opensource
63 likes · 2 comments · 1,473 views
Top 5 projetos open source do GitHub | Vol. 7 #shorts
Os projetos #5 a #1 do Top 10 open source do GitHub, Vol. 7, 10 segundos para cada um.
#5 OpenMAIC: THU-MAIC/OpenMAIC
#4 ManimGX: academa-labs/manimgx
#3 Anatomy: wheresryan22/anatomy
#2 DigUp: ARahim3/DigUp
#1 Doory: shaongitbd/doory
Vídeo completo com os 10 no canal Hallen Tech.
Oferecimento: Hallen Tech, hallentech.com
#shorts #opensource #github
21 likes · 449 views
10 projetos open source do GitHub para testar hoje | Vol. 7
10 projetos open source do GitHub para você testar hoje, em contagem regressiva e com 10 segundos para cada um. Tem alternativa ao Canva e ao CapCut, aula inteira criada por agentes de inteligência artificial, busca no Mac pelo conteúdo dos arquivos, animações de matemática 90 vezes mais rápidas e mais.
📺 Sétimo volume da série Open source do GitHub. Todos com commit nas últimas três semanas (conferido em 14/10/2026). Segue o canal e ativa as notificações para não perder a próxima.
Qual desses você vai testar primeiro? Conta nos comentários!
⏱️ CAPÍTULOS E LINKS
0:00 Abertura
0:10 #10 ChartGPU: gráficos com WebGPU para milhões de pontos em tempo real
ChartGPU/ChartGPU
0:20 #9 Local Dream: Stable Diffusion no celular Android, com a NPU do Snapdragon
xororz/local-dream
0:30 #8 Shaders: efeitos com WebGPU como componentes para React, Vue e Svelte
shader-effects-inc/shaders
0:40 #7 Edit Banana: imagem de diagrama vira arquivo editável do draw.io
BIT-DataLab/Edit-Banana
0:50 #6 Concat: editor de vídeo grátis e open source, alternativa ao CapCut
jub0t/concat
1:00 #5 OpenMAIC: agentes de IA criam uma aula interativa (Universidade Tsinghua)
THU-MAIC/OpenMAIC
1:10 #4 ManimGX: animações no estilo Manim, em Rust, até 90 vezes mais rápidas
academa-labs/manimgx
1:20 #3 Anatomy: skill que explica ideias técnicas com ilustrações isométricas animadas
wheresryan22/anatomy
1:30 #2 DigUp: busca os arquivos do Mac pelo conteúdo, tudo local
ARahim3/DigUp
1:40 Oferecimento Hallen Tech
1:54 #1 Doory: alternativa open source ao Canva, no navegador
shaongitbd/doory
2:04 Encerramento
🧡 Oferecimento: Hallen Tech, liderança de tecnologia sob demanda, software, cloud e segurança para a sua empresa.
Site: hallentech.com
E-mail: contato@hallentech.com
Instagram: instagram.com
LinkedIn da Hallen Tech: linkedin.com
LinkedIn do Eddwan: linkedin.com
Se curtiu, deixa o like, segue o canal e ativa as notificações para a próxima lista.
#opensource #github #inteligenciaartificial
74 likes · 1,818 views
Mentions
GitHub - ARahim3/DigUp: Search your computer by what's in your files: photos, PDFs, documents, recordings, videos and code. On-device on Mac, Windows and Linux, built on Google DeepMind's EmbeddingGemma 2. · GitHub ARahim3/DigUp
Describe the file you lost and DigUp finds it, fully on-device, down to the exact PDF page or video moment. 🔗 ARahim3/DigUp Credits: ARahim3/DigUp #digup #swift #semanticsearch
These Free GitHub Repos Feel Illegal to Use (Clone Viral Videos, Replace Mailchimp) Five free apps that feel like they shouldn't be allowed. One clones any viral video's whole format in one command. One cuts your newsletter bill down to a few dollars of email. One lets you search your own computer like Google, by what's inside your files, without anything leaving your laptop. I installed all five on my own Mac and tested every one on demo files. Here's what works, how long setup really takes, and the catch on each. 📄 FREE: The Loophole List: all five download links, the exact commands I ran, setup times, my test results and every catch, in one PDF. Comment LOOPHOLE and I'll send you the free Loophole List. ▶️ Tip: watch at 1.25× or 1.5× speed for the best experience. ⏱ Chapters 0:00 Five free apps that feel illegal 0:44 How I rate them: works, half works, hype 1:01 #5 Filee: convert any file, offline 2:06 #4 DigUp: search your files by what's in them 3:21 #3 ai-memory: one memory for every AI tool 4:37 #2 MailCheap: the newsletter app that started a fight 6:08 #1 Hypit: clone any viral video 7:26 Copying a format: fair or not? 8:00 The scoreboard 8:35 Your turn 🧰 The 5 free repos • #5 Filee, a drag-and-drop file converter that works offline: KnifeLemon/Filee • #4 DigUp, search your files by what's in them, on your own laptop: ARahim3/DigUp • #3 ai-memory, one shared memory for Claude Code, Codex and 20+ AI tools: akitaonrails/ai-memory • #2 MailCheap, a newsletter app where you pay per email sent: marclou/mailcheap • #1 Hypit, clone any viral video's format with your AI coding agent: hypit-ai/hypit 📊 My results (one Apple Silicon Mac, Oct 25, 2026) • Filee: WORKS. Under 1 min to set up. iPhone photo to JPG 0.51 s, Word to PDF 0.83 s, 3 PDFs merged 0.26 s, with the internet blocked. • DigUp: WORKS. 3 min to set up (865 MB model download). 3 of 3 plain-English searches right on the first try. • ai-memory: WORKS. About 2 min to set up. Codex picked up a Claude Code session, including the failed fix and the missing cent, without me re-explaining. • MailCheap: HALF WORKS. 3 min to the dashboard, Mailchimp import worked, but nothing sends until Amazon SES approves you, and self-hosting has no support. • Hypit: WORKS. A finished clone in 10 min for $0 with code-drawn visuals. AI people or footage need its paid service (about $1.15 a video, plans from $9.90/mo). 🎬 Don't want to touch a coding agent? We built Make Viral Reels: paste any link (or pick a story going viral in your niche), choose a style, and get a finished, ready-to-post reel with voice, captions and music. No face, no editing. Your first video is free: makeviralreels.com Honest notes: • Pricing: Mailchimp at 5,001 to 10,000 contacts starts at $110/mo (Essentials, regular price); beehiiv at 10,000 is $119/mo (Lite, yearly). Both checked logged out on Oct 25, 2026. • Credits: the Filee and DigUp app demo footage is from each project's own README (KnifeLemon/Filee and ARahim3/DigUp, MIT). The tier-list reference clip is Hypit's own example from its README. Tweets shown are public posts by Marc Lou and Tyler Denk. • Every test used made-up demo files in a test folder, never real personal files. One test per project, not a study. 📚 Go deeper with my guides: • The Complete Claude Code Guide: hyperautomationlabs.gumroad.com • The OpenAI Codex Guide: hyperautomationlabs.gumroad.com • Claude Cowork Sales Guide: hyperautomationlabs.gumroad.com • Claude Certified Architect Prep Kit: hyperautomationlabs.gumroad.com ❤️ Support the channel: hit Join on YouTube to become a member, or become a Supporter on Facebook. Deep dives like this live here on YouTube → youtube.com Bite-size AI drops on Instagram → instagram.com Join the community on Facebook → facebook.com Nothing in this video is sponsored. None of these projects paid to be included, and I'm not affiliated with any of them. Make Viral Reels is my own product. #GitHub #OpenSource #FreeApps #ClaudeCode #AI
Ich habe 50+ Github Repos getestet. Das sind die 6 BESTEN (Oktober 2026) 🚀 AUTelligence Solutions (KI-Lösungen für Unternehmen): autelligence-solutions.com 📞 Kostenloser Call mit mir (arbeite mit mir zusammen): cal.flowwy.de 🎓 Meine Community (alle Ressourcen & Templates): skool.com 🔗 Alle GitHub Repos aus dem Video: ▪️ Claude Motion Kit (mein eigenes Repo, einzigartige Motion Graphics mit Claude): nikisge/claude-motion-kit ▪️ Insomnia (Agenten arbeiten bei zugeklapptem MacBook weiter): krishhgg/Insomnia ▪️ Security Audit Skill von Cloudflare (ein Team aus KI-Agenten prüft deinen Code auf Sicherheitslücken): cloudflare/security-audit-skill ▪️ Strata (Qwen 3.8 Flash Next lokal auf deinem Gaming-PC): Niko1221/Strata ▪️ DigUp (deine Dateien nach Inhalt durchsuchen, komplett lokal): ARahim3/DigUp ▪️ God's Eye View (die ganze Welt live auf einem 3D-Globus): bilawalsidhu/gods-eye-view 🛠️ Extras aus dem Video: ▪️ HyperFrames (macht aus HTML ein Video, Basis für das Motion Kit): heygen-com/hyperframes ▪️ What Ships (Launch-Videos als Referenzen für Claude): whatships.com In diesem Video zeige ich dir 6 GitHub Repos, die meine Arbeit mit KI komplett verändert haben – und alle kannst du kostenlos nutzen. Egal ob du mit Claude coole Videos und Motion Graphics erstellen willst, dich für lokale KI interessierst oder viel mit Claude Code und Codex arbeitest: Hier ist für dich etwas dabei. Du siehst: ▪️ wie du mit meinem eigenen Claude Motion Kit und HyperFrames Motion Graphics baust, die nicht nach KI-Slop aussehen (mit echten Referenzvideos, Storyboard und Review-Phase) ▪️ wie Insomnia deine Coding-Agenten weiterarbeiten lässt, während dein MacBook zugeklappt ist ▪️ warum gerade so viele Sicherheitslücken auf einmal auftauchen (PHP, Python, Node.js) und wie der Security Audit Skill von Cloudflare deinen Code von einem ganzen Team aus KI-Agenten prüfen lässt ▪️ wie Strata das 125B-Modell Qwen 3.8 Flash Next schon ab 12 GB Grafikspeicher und 32 GB Arbeitsspeicher lokal laufen lässt – und warum Mixture of Experts das so schnell macht ▪️ wie du mit DigUp deine Videos, Audios, Bilder und PDFs nach Inhalt durchsuchst, komplett lokal auf deinem Mac ▪️ wie du mit God's Eye View Flüge, Verkehr und Live-Kameras auf der ganzen Welt verfolgst – sogar per Sprachsteuerung 📹 Strata im ausführlichen Test bei Fabian: Lokale KI: 125B-Modell auf einer Gaming-Grafikkarte? Strata im Test youtu.be ⏱️ Zeitstempel: 0:00 Einstieg 0:37 1. GitHub Repo 5:04 2. GitHub Repo 6:18 3. GitHub Repo 9:06 4. GitHub Repo 11:39 5. GitHub Repo 13:13 6. GitHub Repo 16:01 Abschluss 💬 Welches der 6 Repos probierst du als Erstes aus? Und wenn du ein ausführliches Video zum Claude Motion Kit willst, schreib es mir in die Kommentare. #ki #github #opensource #claudecode #codex #motiongraphics #hyperframes #lokaleki #kiagenten #cybersecurity #ai #tools #kostenlos [en-US] I Tested 50+ GitHub Repos. These Are the 6 BEST (October 2026) 🚀 AUTelligence Solutions (AI solutions for businesses): autelligence-solutions.com 📞 Free call with me (work with me): cal.flowwy.de 🎓 My community (all resources & templates): skool.com 🔗 All GitHub repositories from the video: ▪️ Claude Motion Kit (my own repository, unique motion graphics with Claude): nikisge/claude-motion-kit ▪️ Insomnia (agents work with a closed MacBook) (continued): krishhgg/Insomnia ▪️ Cloudflare Security Audit Skill (a team of AI agents checks your code for vulnerabilities): cloudflare/security-audit-skill ▪️ Strata (Qwen 3.8 Flash Next locally on your gaming PC): Niko1221/Strata ▪️ DigUp (search your files for content, completely locally): ARahim3/DigUp ▪️ God's Eye View (the whole world live on a 3D globe): bilawalsidhu/gods-eye-view 🛠️ Extras from the video: ▪️ HyperFrames (turns HTML into video, basis for motion) Kit): heygen-com/hyperframes ▪️ What Ships (launch videos as references for Claude): whatships.com In this video, I'll show you six GitHub repositories that have completely transformed my work with AI—and you can use them all for free. Whether you want to create cool videos and motion graphics with Claude, are interested in local AI, or work extensively with Claude Code and Codex: there's something for you here. You'll see: ▪️ how to build motion graphics that don't look like AI slop using my own Claude Motion Kit and HyperFrames (with real reference videos, storyboard, and review phase) ▪️ how Insomnia lets your coding agents keep working while your MacBook is closed ▪️ why so many security vulnerabilities are suddenly appearing (PHP, Python, Node.js) and how Cloudflare's Security Audit Skill lets an entire team of AI agents check your code ▪️ how Strata makes the Qwen 3.8 Flash Next 125B model run locally with as little as 12 GB of graphics memory and 32 GB of RAM – and why Mixture of Experts makes it so fast ▪️ how to search your videos, audio, images, and PDFs by content using DigUp, all locally on your Mac ▪️ how to use God's Eye View to view flights, traffic, and live cameras Tracked worldwide – even via voice control 📹 Strata in an in-depth review by Fabian: Local AI: 125B model on a gaming graphics card? Strata put to the test youtu.be ⏱️ Timestamps: 0:00 Introduction 0:37 1st GitHub Repo 5:04 2nd GitHub Repo 6:18 3rd GitHub Repo 9:06 4th GitHub Repo 11:39 5th GitHub Repo 13:13 6th GitHub Repo 16:01 Conclusion 💬 Which of the 6 repos will you try first? And if you'd like a detailed video about the Claude Motion Kit, let me know in the comments. #AI #GitHub #OpenSource #ClaudeCode #Codex #MotionGraphics #Hyperframes #LocalAI #AIAgents #Cybersecurity #AI #Tools #Free
DigUp Searches Local Media, Documents and Code On-Device DigUp is available as a free, open-source desktop application for searching selected local documents, images, recordings, videos and source code. Its macOS build is accompanied by beta releases for Windows and Linux. For developers, the immediate consequence is one search interface that can locate a function, screenshot, PDF page or point in a recording without uploading those files to a hosted service, according to the project repository. One query, two retrieval paths DigUp runs Google DeepMind’s EmbeddingGemma 2 through llama.cpp. Its readers convert files into text, pictures or sound representations, then store vectors for pages, passages, images, video frames and audio segments in a SQLite index. Those vectors occupy the same 768-dimensional space, allowing one text query to rank different media types together, according to the implementation description. Semantic retrieval is only one part of the search path. DigUp also maintains a keyword index and boosts exact-word matches, which is intended to preserve literal searches for names, invoice numbers and error codes that an embedding may not represent precisely. Queries therefore combine vector proximity with lexical matching rather than treating embeddings as a substitute for exact search, as the README explains. Indexing granularity depends on the format. DigUp processes PDFs page by page, splits audio into overlapping 30-second windows, and samples video frames while also indexing the soundtrack. Search results can point to a PDF page or media timestamp, although jumping directly to that location depends on the viewer available on each operating system, according to the supported-format and playback notes. Code stays in a separate index Code search is disabled until the user enables it for selected folders. It reads source files, notebooks without their outputs and repository documentation, then opens matches at the relevant line in a configured editor. The code index remains separate from general file search, and code queries must begin with code:, according to the code-search documentation. The project says this path skips files ignored by Git, along with vendored or generated code, data files and content that resembles a key or password. It lists support for editors including VS Code, Cursor, Zed, Sublime Text and JetBrains IDEs, plus several platform-specific options, in the same documentation. Local processing still has a cost The maintainers state that indexed content does not leave the computer and that DigUp has no account requirement or telemetry. The application connects to download its model once and to check for updates once per day; the update check can be disabled and is described as sending no information about indexed files. These are claims made by the project itself. That design shifts computation and storage onto the user’s machine. The initial model download is 865 MB, and the first indexing pass can take from minutes to hours depending on the selected files and hardware. DigUp estimates that duration during setup; on Windows and Linux it uses a GPU when a setup check finds that path faster than the processor while producing the same results, according to the installation and performance notes. Platform constraints are uneven. The Mac version requires Apple Silicon and macOS 14 or later. Windows requires 64-bit Windows 10 or 11, and its beta installer is not yet signed. The documented Linux packages target 64-bit Intel and AMD systems, with .deb and AppImage distributions available from the project’s releases and installation instructions. Analysis: broader retrieval means broader access Analysis: DigUp’s consequential design choice is the combination of cross-media vector ranking with exact-string retrieval, not semantic search alone. That can make fragmented project material searchable through one interface, while the separate code index limits accidental mixing between repositories and personal files. The architecture and separation are documented in the repository. The unresolved trade-off is scope. Better retrieval requires granting the application access to more folders and paying the corresponding indexing cost. Folder selection, opt-in code search and exclusions for several sensitive file categories narrow that boundary, but they do not remove the need to decide which local material belongs in a consolidated search index. Those controls and exclusions are described by the project; judging whether they are sufficient remains a user-side trust decision.
DigUp: Describe it. Dig it up. Search your computer by what's in your files (4.6k stars) What it does All text below is quoted directly from the project README. >Search your computer by what's in your files: photos, PDFs, documents, recordings, videos and code. On-device on Mac, Windows and Linux, built on Google DeepMind's EmbeddingGemma 2. What you can find: - "zebra in a video" opens the clip at the moment the zebra shows up - "where they talk about sleep" jumps to that minute of a podcast - "the clause about pets in the lease" shows the PDF page - "a dog on the beach" finds the photo, and "payment declined error" the screenshot - a search in English can find a note written in Bengali, Arabic or any of the model's 100+ languages, and the other way around - "code: retry a failed call with backoff" finds the function in your projects, if you turn on code search >Nothing you index leaves your computer. DigUp goes online only to download the model once and to check for updates once a day. The update check sends nothing about your files, and you can turn it off. No account, no telemetry. >Mac. Download the DMG from Releases and drag DigUp to Applications, or use Homebrew: brew install --cask arahim3/tap/digup. It needs an Apple Silicon Mac with macOS 14 or later. Repository ARahim3/DigUp Repository stats |Metric|Value| |:-|:-| |Stars|4,631| |Forks|333| |Last release|v0.6.1 (2026-10-12)| |Languages|Swift 66.5%, Rust 12.3%, JavaScript 8.7%| |Contributors|4| |AI contributor accounts (Claude / Codex / Cursor / Copilot)|None| |AI co-authored commits|None| Main contributor |Field|Value| |:-|:-| |Name|Abdur Rahim| |GitHub|@ARahim3| |Commits to this repo|142 (top contributor)| |X / Twitter|@ARahim| |Followers|233|
No telemetry. It's free and open source. 🔗 ARahim3/DigUp Imagine never hunting through folders again. #AI #OpenSource #Productivity #DeveloperTools
Repos de GitHub para la IA: top 3 de esta semana Tres repos de GitHub de esta semana si usas la IA: una skill que te explica cosas construyéndolas, un buscador que encuentra el segundo exacto de un vídeo y una placa de 7 $ con la que una IA usa tu iPhone. Anatomy: wheresryan22/anatomy DigUp: ARahim3/DigUp ESP32 iPhone Agent: onehumanbeing/esp32-iphone-agent #github #ia #claudecode #opensource
Top 5 projetos open source do GitHub | Vol. 7 #shorts Os projetos #5 a #1 do Top 10 open source do GitHub, Vol. 7, 10 segundos para cada um. #5 OpenMAIC: THU-MAIC/OpenMAIC #4 ManimGX: academa-labs/manimgx #3 Anatomy: wheresryan22/anatomy #2 DigUp: ARahim3/DigUp #1 Doory: shaongitbd/doory Vídeo completo com os 10 no canal Hallen Tech. Oferecimento: Hallen Tech, hallentech.com #shorts #opensource #github
10 projetos open source do GitHub para testar hoje | Vol. 7 10 projetos open source do GitHub para você testar hoje, em contagem regressiva e com 10 segundos para cada um. Tem alternativa ao Canva e ao CapCut, aula inteira criada por agentes de inteligência artificial, busca no Mac pelo conteúdo dos arquivos, animações de matemática 90 vezes mais rápidas e mais. 📺 Sétimo volume da série Open source do GitHub. Todos com commit nas últimas três semanas (conferido em 14/10/2026). Segue o canal e ativa as notificações para não perder a próxima. Qual desses você vai testar primeiro? Conta nos comentários! ⏱️ CAPÍTULOS E LINKS 0:00 Abertura 0:10 #10 ChartGPU: gráficos com WebGPU para milhões de pontos em tempo real ChartGPU/ChartGPU 0:20 #9 Local Dream: Stable Diffusion no celular Android, com a NPU do Snapdragon xororz/local-dream 0:30 #8 Shaders: efeitos com WebGPU como componentes para React, Vue e Svelte shader-effects-inc/shaders 0:40 #7 Edit Banana: imagem de diagrama vira arquivo editável do draw.io BIT-DataLab/Edit-Banana 0:50 #6 Concat: editor de vídeo grátis e open source, alternativa ao CapCut jub0t/concat 1:00 #5 OpenMAIC: agentes de IA criam uma aula interativa (Universidade Tsinghua) THU-MAIC/OpenMAIC 1:10 #4 ManimGX: animações no estilo Manim, em Rust, até 90 vezes mais rápidas academa-labs/manimgx 1:20 #3 Anatomy: skill que explica ideias técnicas com ilustrações isométricas animadas wheresryan22/anatomy 1:30 #2 DigUp: busca os arquivos do Mac pelo conteúdo, tudo local ARahim3/DigUp 1:40 Oferecimento Hallen Tech 1:54 #1 Doory: alternativa open source ao Canva, no navegador shaongitbd/doory 2:04 Encerramento 🧡 Oferecimento: Hallen Tech, liderança de tecnologia sob demanda, software, cloud e segurança para a sua empresa. Site: hallentech.com E-mail: contato@hallentech.com Instagram: instagram.com LinkedIn da Hallen Tech: linkedin.com LinkedIn do Eddwan: linkedin.com Se curtiu, deixa o like, segue o canal e ativa as notificações para a próxima lista. #opensource #github #inteligenciaartificial