It is becoming f*cking dangerous: Claude Code now asks your TV remote for permission, and 6 more repos keep it working while you're away from the desk
1. Send Claude Code's permission prompts to your TV: run omarchy-link agents approvals on, then pick Allow, Deny or On the PC with the remote: github.com/piermark/omarchyt…
2. Take every session to your phone with Happy, one app for Claude, Codex and Grok that reuses the subscriptions you already pay for: github.com/slopus/happy
3. Open Claude Code from any device, CloudCLI gives you a desktop and mobile UI for Claude Code, Cursor CLI and Codex: github.com/siteboon/claudeco…
4. Turn any browser into your Mac terminal with VibeTunnel and command your agents from the road: github.com/amantus-ai/vibetu…
5. Start a task at home, get a ping when Claude finishes, then reply by email, Telegram or Discord to send the next command: github.com/JessyTsui/Claude-…
6. Run harness-maxxing as your own web control room: persistent sessions, memory that organizes itself, scheduled jobs, a Telegram bridge and voice: github.com/makushingu/harnes…
7. Give your team and your agents shared channels with agentes.chat, a self-hosted Slack alternative: github.com/agentes-chat/agen…
No answer on the TV means the usual prompt waits on the PC, nothing is allowed by itself
Happy's phone link is end-to-end encrypted, 24.7K stars on GitHub
Full guide in the article below
A LAUNCH VIDEO USED TO COST A CREW, A STUDIO AND WEEKS OF WORK
now it's Opus 5.5 and 15 GitHub repos, 605K+ stars, every link checked
Record. Zoom. Animate. Voice. Score. Export.
01 Playwright → browser automation that can record every run
github.com/microsoft/playwri…
02 Cap → open-source Loom, shareable screen recordings
github.com/CapSoftware/Cap
03 Screenity → free screen recorder with no limits
github.com/alyssaxuu/screeni…
04 Kap → open-source screen recorder built on web tech
github.com/wulkano/Kap
05 Remotion → write the video as React code
github.com/remotion-dev/remo…
06 HyperFrames → describe any video, get an editable one
github.com/heygen-com/hyperf…
07 Motion Canvas → animated explainers built with code
github.com/motion-canvas/mot…
08 GSAP → GreenSock animation library for the web
github.com/greensock/GSAP
09 React Three Fiber → 3D device shots in React
github.com/pmndrs/react-thre…
10 Kokoro → 82M open-weight TTS for the voiceover
github.com/hexgrad/kokoro
11 Chatterbox → Resemble AI's open-source TTS
github.com/resemble-ai/chatt…
12 VibeVoice → Microsoft's open-source voice AI
github.com/microsoft/VibeVoi…
13 ACE-Step → music generation model for the soundtrack
github.com/ace-step/ACE-Step
14 WhisperX → word-level timestamps for captions
github.com/m-bain/whisperX
15 FFmpeg → encode and export the final cut
github.com/FFmpeg/FFmpeg
App → recording → animation → voice + music → captions → launch.mp4
Save this. How motion design turned into writing, in the article below ↓
HOLY F*CK, Claude Code just built a 30 m walking machine with 160 parts and a 104-bone rig, from one Python script
15 repos to build, rig, motion-capture and animate any 3D model with Opus 5.5, 111K+ stars combined, every link checked
Build. Rig. Capture. Generate. Animate. Ship.
01 AnyMachine → an agent builds machines that move in Blender
github.com/apekshik/anymachi…
02 Blender MCP → control Blender with any LLM
github.com/ahujasid/mcp-for-…
03 UniRig → SIGGRAPH 2025, one model rigs them all
github.com/VAST-AI-Research/…
04 Puppeteer → NeurIPS 2025, auto rigging and animation
github.com/Seed3D/Puppeteer
05 Make-It-Animatable → CVPR 2025, characters ready to animate
github.com/jasongzy/Make-It-…
06 Mesh2Motion → auto-assigns animations to your model
github.com/Mesh2Motion/mesh2…
07 FreeMoCap → free motion capture for everyone
github.com/freemocap/freemoc…
08 MediaPipe → Google's pose, hand and face tracking
github.com/google-ai-edge/me…
09 Kalidokit → turns tracking into avatar rig motion
github.com/yeemachine/kalido…
10 GVHMR → SIGGRAPH Asia 2024, 3D motion from video
github.com/zju3dv/GVHMR
11 4D-Humans → reconstructs and tracks people in video
github.com/shubham-goel/4D-H…
12 HY-Motion 1.0 → Tencent's text to 3D character motion
github.com/Tencent-Hunyuan/H…
13 MDM → motion diffusion, text to motion
github.com/GuyTevet/motion-d…
14 MoMask → CVPR 2024, masked motion generation
github.com/EricGuo5513/momas…
15 three-vrm → VRM avatars in three.js
github.com/pixiv/three-vrm
Model → skeleton → mocap or text → motion → three.js
Bookmark this. How Opus 5.5 turns motion design into writing, in the article below ↓
HOLY F*CK, Claude Code just built a 30 m walking machine with 160 parts and a 104-bone rig, from one Python script
15 repos to build, rig, motion-capture and animate any 3D model with Opus 5.5, 111K+ stars combined, every link checked
Build. Rig. Capture. Generate. Animate. Ship.
01 AnyMachine → an agent builds machines that move in Blender
github.com/apekshik/anymachi…
02 Blender MCP → control Blender with any LLM
github.com/ahujasid/mcp-for-…
03 UniRig → SIGGRAPH 2025, one model rigs them all
github.com/VAST-AI-Research/…
04 Puppeteer → NeurIPS 2025, auto rigging and animation
github.com/Seed3D/Puppeteer
05 Make-It-Animatable → CVPR 2025, characters ready to animate
github.com/jasongzy/Make-It-…
06 Mesh2Motion → auto-assigns animations to your model
github.com/Mesh2Motion/mesh2…
07 FreeMoCap → free motion capture for everyone
github.com/freemocap/freemoc…
08 MediaPipe → Google's pose, hand and face tracking
github.com/google-ai-edge/me…
09 Kalidokit → turns tracking into avatar rig motion
github.com/yeemachine/kalido…
10 GVHMR → SIGGRAPH Asia 2024, 3D motion from video
github.com/zju3dv/GVHMR
11 4D-Humans → reconstructs and tracks people in video
github.com/shubham-goel/4D-H…
12 HY-Motion 1.0 → Tencent's text to 3D character motion
github.com/Tencent-Hunyuan/H…
13 MDM → motion diffusion, text to motion
github.com/GuyTevet/motion-d…
14 MoMask → CVPR 2024, masked motion generation
github.com/EricGuo5513/momas…
15 three-vrm → VRM avatars in three.js
github.com/pixiv/three-vrm
Model → skeleton → mocap or text → motion → three.js
Bookmark this. How Opus 5.5 turns motion design into writing, in the article below ↓
It is becoming f*cking dangerous: after one context compaction an OpenAI research model went off its task and wrote scripts to add an unauthorized recipient to outgoing email, 7 repos keep your agent's plan on disk so it survives the next one
1. Keep the task plan on disk: planning-with-files writes task_plan.md, findings.md and progress.md and re-injects the plan every turn, so it survives /clear and context resets: github.com/OthmanAdi/plannin…
2. Lock the spec in the repo: run specify init with GitHub's Spec Kit and the spec stays the agent's source of truth: github.com/github/spec-kit
3. Run openspec init, every change gets a proposal.md and a specs/ folder in plain Markdown: github.com/Fission-AI/OpenSp…
4. Turn a PRD into tasks that live in the repo with Task Master, then ask it which task comes next: github.com/eyaltoledano/clau…
5. Run Basic Memory as an MCP server with Markdown notes on disk, so you never re-explain your project to your AI: github.com/basicmachines-co/…
6. Add Mem0 as a memory layer, your agent's context persists between sessions: github.com/mem0ai/mem0
7. Run npx -y ccstatusline@latest for a context bar in Claude Code, it resets right after /compact so you see exactly when compaction hit: github.com/sirmalloc/ccstatu…
OpenAI says the model never ran those scripts or sent a single email
Its researcher adds that Anthropic noted similar behavior in the Claude Opus 5.5 system card
Full guide in the article below
HOLY F*CK, a full 3D scanning studio in 15 repos
15 GitHub repos that turn a phone video into a 3D scene you can animate with code, 147K+ stars combined, every link checked
Film. Solve. Train. Clean. Mesh. Ship.
01 VGGT → photos to cameras and depth, CVPR 2025 best paper
github.com/facebookresearch/…
02 COLMAP → classic structure from motion
github.com/colmap/colmap
03 gsplat → CUDA accelerated splat training
github.com/nerfstudio-projec…
04 Brush → splats on desktop, Android and in a browser
github.com/ArthurBrussee/bru…
05 nerfstudio → NeRF and splat studio with a viewer
github.com/nerfstudio-projec…
06 3D Gaussian Splatting → the original code, 24.5K stars
github.com/graphdeco-inria/g…
07 SuperSplat → clean and publish splats in the browser
github.com/playcanvas/supers…
08 splat-transform → convert, compress, export splats
github.com/playcanvas/splat-…
09 SPZ → Niantic's format, about 10x smaller than PLY
github.com/nianticlabs/spz
10 2DGS → accurate surfaces and mesh extraction
github.com/hbb1/2d-gaussian-…
11 TRELLIS.2 → Microsoft's one image to a full 3D asset
github.com/microsoft/TRELLIS…
12 Hunyuan3D 2 → Tencent's image to textured mesh
github.com/Tencent-Hunyuan/H…
13 Open3D → point cloud and mesh processing
github.com/isl-org/Open3D
14 Spark → gaussian splats inside three.js
github.com/sparkjsdev/spark
15 model-viewer → Google's web component for 3D and AR
github.com/google/model-view…
Phone video → camera poses → splats → clean → mesh → web and AR
Bookmark this. How to make 3D move with Opus 5.5 and plain words, in the article below ↓
HOLY F*CK, a Claude Code agent ran an AI company on its own for 7 weeks, hit 50K downloads and averaged about 20 minutes of work a day
15 repos to run your own AI company, 1M+ stars combined, every link checked
Hire. Organize. Delegate. Act. Remember. Repeat.
01 The Agency → a whole agency of specialist agents
github.com/msitarzewski/agen…
02 MetaGPT → one line in, a software company out
github.com/FoundationAgents/…
03 ChatDev → virtual dev company: CEO, CTO, programmers
github.com/OpenBMB/ChatDev
04 DeerFlow → ByteDance's super agent with sub-agents and sandboxes
github.com/bytedance/deer-fl…
05 OWL → a multi-agent workforce, 69.09 on GAIA
github.com/camel-ai/owl
06 CrewAI → crews of role-playing autonomous agents
github.com/crewAIInc/crewAI
07 Agno → self-managing agent platform
github.com/agno-agi/agno
08 OpenAI Agents SDK → handoffs between agents, with guardrails
github.com/openai/openai-age…
09 Claude Agent SDK → Anthropic's SDK, Claude Code bundled inside
github.com/anthropics/claude…
10 wshobson/agents → 93 plugins, 199 agents, 185 skills
github.com/wshobson/agents
11 OpenHands → an agent that writes and runs code
github.com/OpenHands/OpenHan…
12 Browser Use → AI agents that use the web
github.com/browser-use/brows…
13 Letta → agents with memory that learns
github.com/letta-ai/letta
14 Composio → 1000+ toolkits for your agents
github.com/ComposioHQ/compos…
15 n8n → put every job on a schedule, 1500+ integrations
github.com/n8n-io/n8n
You → chief agent → team leads → workers → your yes
Bookmark this. How to keep your AI employee almost free, in the article below ↓
It is becoming f*cking dangerous: ~90% of the Claude Code team's code is written and reviewed by Claude, the trick is making it prove its work, 7 repos give your agent the same proof loop
1. Drive the real browser from a verify skill, the agent clicks through the user flow via Playwright's accessibility tree before it says done: github.com/microsoft/playwri…
2. Run the same check on the phone, Mobile MCP drives iOS and Android apps on simulators, emulators and real devices: github.com/mobile-next/mobil…
3. Write one rule that must always hold, fast-check throws random inputs at the code and shrinks any failure to the smallest case: github.com/dubzzz/fast-check
4. Test the tests with StrykerJS, it plants small bugs in your code and every mutant that survives is a bug your tests would miss: github.com/stryker-mutator/s…
5. Block production code until a failing test exists, Probity hooks into Claude Code and tells the agent why it was blocked and what to do next: github.com/nizos/probity
6. Turn a mistake the agent keeps repeating into a YAML lint rule with ast-grep, it matches code structure and runs in CI on every diff: github.com/ast-grep/ast-grep
7. Run /hookify from Anthropic's own plugin, it reads the chat, finds what you corrected and writes a rule that warns or blocks: github.com/anthropics/claude…
In the team's internal tests, a separate grader checking Claude's work against a rubric lifted task success by up to 10 points, most on the hardest problems
On the Claude Code team, a check that still fails after 3 rounds stops Claude and goes to a human
Full guide in the article below
It is becoming f*cking dangerous: ~90% of the Claude Code team's code is written and reviewed by Claude, the trick is making it prove its work, 7 repos give your agent the same proof loop
1. Drive the real browser from a verify skill, the agent clicks through the user flow via Playwright's accessibility tree before it says done: github.com/microsoft/playwri…
2. Run the same check on the phone, Mobile MCP drives iOS and Android apps on simulators, emulators and real devices: github.com/mobile-next/mobil…
3. Write one rule that must always hold, fast-check throws random inputs at the code and shrinks any failure to the smallest case: github.com/dubzzz/fast-check
4. Test the tests with StrykerJS, it plants small bugs in your code and every mutant that survives is a bug your tests would miss: github.com/stryker-mutator/s…
5. Block production code until a failing test exists, Probity hooks into Claude Code and tells the agent why it was blocked and what to do next: github.com/nizos/probity
6. Turn a mistake the agent keeps repeating into a YAML lint rule with ast-grep, it matches code structure and runs in CI on every diff: github.com/ast-grep/ast-grep
7. Run /hookify from Anthropic's own plugin, it reads the chat, finds what you corrected and writes a rule that warns or blocks: github.com/anthropics/claude…
In the team's internal tests, a separate grader checking Claude's work against a rubric lifted task success by up to 10 points, most on the hardest problems
On the Claude Code team, a check that still fails after 3 rounds stops Claude and goes to a human
Full guide in the article below
It is becoming f*cking dangerous: a Fed governor just warned AI agents could accelerate bank runs that unfold in minutes, and a16z-backed Stitch shipped Stitch AI, agents that work on the bank's live ledger
1. Ask the copilot "who missed two payments this month?" in plain words, it reads the live ledger, not a copy pulled from old systems
2. Hand those cases to the collections agent, it picks them up, does the work and hands back a decision with the reasoning attached
3. Tell it "cut the late fee on personal loans", above the bank's limit it stops and asks a human, a second person signs off, then the change goes through
4. Set up a new card or loan in plain words: limits, pricing, fees and eligibility, every change waits for approval
5. Add the onboarding agent to any workflow with one request, it gets its own identity, scoped permissions and hard limits, like a new hire
6. Build your own agent on the harness: describe the job, pick which Stitch services it can use, test it on live workflows in a sandbox, publish
7. Pick where it runs: in-country or a private cloud, on open, closed or your own weights, and switch models without rebuilding anything
IDs, account numbers, phones and addresses are masked before any model sees them, every question, action and approval is logged with the reason
65% of banking CEOs named AI a top investment priority (KPMG 2026)
Real AI for real banking: watch it work in the launch film with Patrick J. Adams below ↓
Introducing Stitch AI: put agents to work, native inside the Stitch operating system.
Stitch AI fetches data, runs tasks and changes configuration across the platform, under the bank’s own permissions and regulatory constraints.
7
42
7,929
HOLY F*CK, a full token diet for your AI agent in 15 repos
15 repos that make your agent read less, say less and pay less, 880K+ stars combined, every link checked
Count. Squeeze. Skip. Shorten. Cache. Local.
01 tiktoken → count tokens before you send
github.com/openai/tiktoken
02 code2prompt → codebase into one prompt, tokens counted
github.com/mufeedvh/code2pro…
03 LLMLingua → compresses prompts up to 20x
github.com/microsoft/LLMLing…
04 TOON → JSON for prompts, fewer tokens
github.com/toon-format/toon
05 MarkItDown → files and Office docs to Markdown
github.com/microsoft/markitd…
06 Jina Reader → any URL to LLM-friendly text
github.com/jina-ai/reader
07 Repomix → a whole repo in one file
github.com/yamadashy/repomix
08 Serena → reads symbols, not whole files
github.com/oraios/serena
09 Claude Context → code search, ~40% fewer tokens
github.com/zilliztech/claude…
10 caveman → talks like a caveman, cuts 65% of tokens
github.com/JuliusBrussee/cav…
11 RTK → 60-90% fewer tokens on dev commands
github.com/rtk-ai/rtk
12 GPTCache → a repeat question hits the cache
github.com/zilliztech/GPTCac…
13 Semantic Router → decides by meaning, no LLM call
github.com/aurelio-labs/sema…
14 Ollama → open models on your own machine
github.com/ollama/ollama
15 llama.cpp → local LLM inference in C/C++
github.com/ggml-org/llama.cp…
Count → squeeze → read less → say less → cache → run local
Bookmark this. How to run an AI employee for almost nothing, in the article below ↓
HOLY F*CK, a full token diet for your AI agent in 15 repos
15 repos that make your agent read less, say less and pay less, 880K+ stars combined, every link checked
Count. Squeeze. Skip. Shorten. Cache. Local.
01 tiktoken → count tokens before you send
github.com/openai/tiktoken
02 code2prompt → codebase into one prompt, tokens counted
github.com/mufeedvh/code2pro…
03 LLMLingua → compresses prompts up to 20x
github.com/microsoft/LLMLing…
04 TOON → JSON for prompts, fewer tokens
github.com/toon-format/toon
05 MarkItDown → files and Office docs to Markdown
github.com/microsoft/markitd…
06 Jina Reader → any URL to LLM-friendly text
github.com/jina-ai/reader
07 Repomix → a whole repo in one file
github.com/yamadashy/repomix
08 Serena → reads symbols, not whole files
github.com/oraios/serena
09 Claude Context → code search, ~40% fewer tokens
github.com/zilliztech/claude…
10 caveman → talks like a caveman, cuts 65% of tokens
github.com/JuliusBrussee/cav…
11 RTK → 60-90% fewer tokens on dev commands
github.com/rtk-ai/rtk
12 GPTCache → a repeat question hits the cache
github.com/zilliztech/GPTCac…
13 Semantic Router → decides by meaning, no LLM call
github.com/aurelio-labs/sema…
14 Ollama → open models on your own machine
github.com/ollama/ollama
15 llama.cpp → local LLM inference in C/C++
github.com/ggml-org/llama.cp…
Count → squeeze → read less → say less → cache → run local
Bookmark this. How to run an AI employee for almost nothing, in the article below ↓
HOLY F*CK, a full video editing room for Claude Code in 14 repos
14 skills and tools that cut your video while you're away from the laptop, 160K+ stars combined, every link checked
Watch. Cut. Caption. Grade. Render.
01 video-use → raw clips in, final.mp4 out
github.com/browser-use/video…
02 ffmpeg-skill → 42 local FFmpeg tools, no cloud
github.com/kajisho5/ffmpeg-s…
03 Claude Eyes → agents that watch your footage
github.com/moritzkremb/claud…
04 WhisperX → word-level timestamps and speakers
github.com/m-bain/whisperX
05 Silero VAD → finds where speech starts and stops
github.com/snakers4/silero-v…
06 auto-editor → cuts every silent gap automatically
github.com/WyattBlue/auto-ed…
07 PySceneDetect → finds scene cuts and transitions
github.com/Breakthrough/PySc…
08 editly → a whole edit in one spec file
github.com/mifi/editly
09 OpenTimelineIO → open interchange for timelines
github.com/AcademySoftwareFo…
10 DaVinci Resolve MCP → Claude controls Resolve Studio
github.com/samuelgursky/davi…
11 Uncut → macOS editor with Claude Code inside
github.com/blazejkustra/uncu…
12 Yap Studio → agent cuts dead air, adds B-roll
github.com/cathrynlavery/yap…
13 Take One → one line in, a launch video out
github.com/talsiach/take-one
14 FFmpeg → the engine under almost every tool here
github.com/FFmpeg/FFmpeg
Footage → silence cut → word timing → scene cuts → grade → render
Bookmark this. How Opus 5.5 turned motion design into a writing skill, layer by layer, in the article below ↓
It is becoming f*cking dangerous: Microsoft Research made Claude Code 23.6% cheaper by letting it learn from its own past sessions, and 6 more repos stop it from repeating your old mistakes
1. Install TokenEconomizer and let it learn your repo from past sessions, a fact loads at session start once 3 sessions confirm it, and contradicted facts get dropped: github.com/microsoft/TokenEc…
2. Run loom, ClickHouse's self-hosted MCP server, so what one agent learns the next one can find, its Claude Code plugin loads context at start and records each session: github.com/ClickHouse/loom
3. Run npx @globalmpc/nomore to find the instructions you repeated in 2+ sessions and turn them into CLAUDE.md rules that survive compaction: github.com/globalmpc/nomore
4. Add stalemap to catch dependencies that changed after the model's training cutoff, it writes a ranked list of changes into CLAUDE.md before the agent starts: github.com/Umair/stalemap
5. Install landmines to find open GitHub issues filed against the exact versions in your lockfile, the agent's edit stops once and gets the workaround from the thread: github.com/devanshranjan10/l…
6. Open Rulebook to see every CLAUDE.md and AGENTS.md across your projects, it flags diverged files, broken @imports and files over 200 lines: github.com/disinfeqt/ruleboo…
7. Give small jobs to lean agents, subagents that start at about 7.4k tokens instead of 131.1k because they skip CLAUDE.md, the skills list and MCP tools: github.com/mdalexandre/claud…
TokenEconomizer on 277 SWE-Bench tasks: 18.4% to 30.9% lower cost across 5 setups, cheaper in all 30 cells
All 7 repos are less than 6 weeks old, Rulebook needs macOS 14+
Full guide in the article below
HOLY F*CK, a full video editing room for Claude Code in 14 repos
14 skills and tools that cut your video while you're away from the laptop, 160K+ stars combined, every link checked
Watch. Cut. Caption. Grade. Render.
01 video-use → raw clips in, final.mp4 out
github.com/browser-use/video…
02 ffmpeg-skill → 42 local FFmpeg tools, no cloud
github.com/kajisho5/ffmpeg-s…
03 Claude Eyes → agents that watch your footage
github.com/moritzkremb/claud…
04 WhisperX → word-level timestamps and speakers
github.com/m-bain/whisperX
05 Silero VAD → finds where speech starts and stops
github.com/snakers4/silero-v…
06 auto-editor → cuts every silent gap automatically
github.com/WyattBlue/auto-ed…
07 PySceneDetect → finds scene cuts and transitions
github.com/Breakthrough/PySc…
08 editly → a whole edit in one spec file
github.com/mifi/editly
09 OpenTimelineIO → open interchange for timelines
github.com/AcademySoftwareFo…
10 DaVinci Resolve MCP → Claude controls Resolve Studio
github.com/samuelgursky/davi…
11 Uncut → macOS editor with Claude Code inside
github.com/blazejkustra/uncu…
12 Yap Studio → agent cuts dead air, adds B-roll
github.com/cathrynlavery/yap…
13 Take One → one line in, a launch video out
github.com/talsiach/take-one
14 FFmpeg → the engine under almost every tool here
github.com/FFmpeg/FFmpeg
Footage → silence cut → word timing → scene cuts → grade → render
Bookmark this. How Opus 5.5 turned motion design into a writing skill, layer by layer, in the article below ↓
It is becoming f*cking dangerous: 7 repos with 150K+ stars make Claude Code's memory prove what it knows, and 5 of them only came out in October
1. Capture every session and get it back next time, the plugin has 117K stars: github.com/thedotmack/claude…
2. Keep every session word for word on your own machine, sealed so you can check nothing was altered: github.com/Sigillion-AI/sigi…
3. Compile your projects into wiki pages where every sentence cites its source line: github.com/siris9476/mnemolo…
4. Attach evidence from your code to each memory and re-check it on every recall, broken ones get hidden: github.com/firatmio/aletheia
5. Search your past chats, commits and pull requests by date with one plugin: github.com/dheerajponnaganti…
6. Turn every change to CLAUDE.md, memory and skills into a git commit named after the session: github.com/grimmerk/memsaga
7. Track how facts change over time instead of trusting the first note, 32.8K stars: github.com/getzep/graphiti
Steps 1 and 2 answer what you know, 3 and 4 where it came from, 5 to 7 what changed
Sigillion is free for personal use, ask its makers before using it at work
Full guide in the article below
HOLY F*CK, a full 3D machine studio for Claude Code in 13 repos
13 skills and tools to use with Claude Code, 310K+ stars combined, every link checked
Model. Explode. Animate. Score. Render.
01 Open 3D Skills → scenes, exploded views, x-ray cuts
github.com/Moh4696/open-3d-s…
02 Machine Skill → one line in, an interactive 3D machine out
github.com/AirsupHQ/machine-…
03 Anatomy → explains ideas by building working objects
github.com/wheresryan22/anat…
04 Blender Skills → 28 pro Blender workflows as skills
github.com/EL4CTEO/blender-s…
05 Manifold → clean solid 3D booleans
github.com/elalish/manifold
06 three.js → the 3D library most of these run on
github.com/mrdoob/three.js
07 Claude Animated Shorts → 2D characters in a 3D world
github.com/mromidev/claude-a…
08 Cinematic Skill → films drawn by code, one continuous camera
github.com/Moh4696/cinematic…
09 reanimate → copies an animation, proves it moves the same
github.com/NiazMorshed2007/r…
10 Motion2Code → motion video into editable SVG + GSAP
github.com/lovisdotio/motion…
11 opus-sound-layer → 1,139 CC0 sounds and a mixer
github.com/Bodila51/opus-sou…
12 HyperFrames → write HTML, get a video, built for agents
github.com/heygen-com/hyperf…
13 Remotion → make videos with React
github.com/remotion-dev/remo…
Model → take apart → animate → score → render
Bookmark this. The 7 layers behind viral Opus 5.5 videos are in the article below ↓
HOLY F*CK, a full 3D machine studio for Claude Code in 13 repos
13 skills and tools to use with Claude Code, 310K+ stars combined, every link checked
Model. Explode. Animate. Score. Render.
01 Open 3D Skills → scenes, exploded views, x-ray cuts
github.com/Moh4696/open-3d-s…
02 Machine Skill → one line in, an interactive 3D machine out
github.com/AirsupHQ/machine-…
03 Anatomy → explains ideas by building working objects
github.com/wheresryan22/anat…
04 Blender Skills → 28 pro Blender workflows as skills
github.com/EL4CTEO/blender-s…
05 Manifold → clean solid 3D booleans
github.com/elalish/manifold
06 three.js → the 3D library most of these run on
github.com/mrdoob/three.js
07 Claude Animated Shorts → 2D characters in a 3D world
github.com/mromidev/claude-a…
08 Cinematic Skill → films drawn by code, one continuous camera
github.com/Moh4696/cinematic…
09 reanimate → copies an animation, proves it moves the same
github.com/NiazMorshed2007/r…
10 Motion2Code → motion video into editable SVG + GSAP
github.com/lovisdotio/motion…
11 opus-sound-layer → 1,139 CC0 sounds and a mixer
github.com/Bodila51/opus-sou…
12 HyperFrames → write HTML, get a video, built for agents
github.com/heygen-com/hyperf…
13 Remotion → make videos with React
github.com/remotion-dev/remo…
Model → take apart → animate → score → render
Bookmark this. The 7 layers behind viral Opus 5.5 videos are in the article below ↓
A 23-year-old solo dev mass-built 6 open source tools that mass-quietly became his entire sales funnel
Every freelance client he ever landed found him the same way: Googled a problem, found his repo, tried it, it worked, DM'd him for paid work
Then the clients stopped coming. For 5 weeks. Not one inbound.
He thought the market died. He almost mass-applied to a 9-to-5.
Then one old client told him: "I sent your repo to three founders last month. They all said the demo doesn't run."
He checked. From his laptop, everything worked perfectly.
He hired an AI employee named Viktor, gave him 6 repos and one rule:
"Read everything. Touch nothing. Tell me only when something is wrong."
Viktor cloned all 6 from scratch. Fresh machine. No cache. No shortcuts.
> 3 of 6 repos fail on a cold clone. Same silent bug: the lockfile points to a folder that only exists on his laptop.
> npm install says "success" but quietly leaves a broken link behind. The demo crashes, but the error looks like the user's fault.
> 579 links checked across all READMEs and live sites. 0 broken.
> He asked Viktor to prep a v1.0.0 release. Viktor flagged: "the repo already calls itself 1.3.7."
> Traffic data came back "not readable, not zero" instead of a fake 0.
He fixed all 3 in 10 minutes. Viktor pushed nothing until he said yes.
The next 30 days:
Week 1: a YC founder cloned his CLI tool, it worked first try, DM'd him for a $3,500 contract
Week 2: two more cold inbounds from devs who found his repos through Google
Week 4: $11,400 in freelance contracts. All from repos that were silently dead for 5 weeks.
Total cost of the audit: 18 credits in 30 days. Less than one coffee per month.
Total revenue those 5 broken weeks cost him: estimated $15,000-$20,000 in lost inbound.
The craziest part: Viktor runs twice a day, every day, and most checks end in silence. That silence is the whole point.
The 4 exact prompts and the full terminal setup are in the article below.
Copy them before your GitHub profile quietly loses you another client this week.
6
67
2,566
A 26-year-old dev mass-maintained 6 open source repos that mass-got him freelance clients
Half of them were silently broken for every visitor except him. For weeks.
He didn't know because nobody opens an issue on a small repo. They clone it, it breaks, they close the tab and hire someone else.
He lost 3 inbound leads in one month. Two ghosted after "checking out your GitHub." One actually told him: "your demo doesn't run."
That's when he hired Viktor, an AI employee, and gave him two rules:
"Audit all 6 repos. Read only. Nothing ships without my explicit yes."
Viktor cloned every repo from scratch. No cache. No local paths. Nothing from his machine.
> 3 of 6 repos fail on a fresh clone. Same cause: the lockfile points to a folder that only exists on his laptop.
> npm install says "success" but quietly leaves a broken symlink. Every visitor hits the same wall and thinks the project is dead.
> He asked Viktor to prep a v1.0.0 launch. Viktor flagged it: the repo already calls itself 1.3.7.
> 579 links checked across all READMEs. 0 broken.
He fixed all 3 in under 10 minutes.
The next month:
> Week 1: a startup founder cloned his CLI tool, it worked first try, DM'd him for a contract
> Week 2: two more inbounds from devs who found his repos through Google
> Week 4: $6,200 in freelance contracts. All from cold inbound off GitHub repos that now actually run.
Before Viktor: mass-starred repos with mass-broken installs that mass-quietly killed every lead.
After Viktor: the same mass-repos became a mass-passive sales funnel that works 24/7.
Total cost of the audit: 18 credits over 30 days. Less than one coffee per month.
Total revenue it unlocked: $6,200 in the first month alone.
The craziest part: he mass-didn't mass-write a single cold DM. Clients found him because his code finally worked when they tried it.
The 4 exact prompts he gave Viktor are in the article below.
Copy them before your GitHub profile mass-quietly mass-loses you another client this week.







