Zephyrx

You can call me mr.zepp

[konstantinpaulus @konstipaulus]

Launching text-to-lottie v1.0.0 (stable)

A framework for generating production ready Lottie animations with Claude Code and Codex.

What’s new:
- Multi-project, multi-scene support
- Drag-and-drop Lottie file import
- Complete UI rewrite

Now at 2.8k GitHub stars (MIT licensed)

Try it: $ npx skills add diffusionstudio/lottie

[Eren Chen @ErenChenAI]

MediaPipe + Python.

I've got a great idea 😅

[Crunio | CGメディア @Botan_cr]

WebGLとThree.jsを活用してWebサイト上で滑らかなゲーム体験をブラウザ完結しています。

メモリは置いておいて、オンラインショッピング、都市探検（旅行サイト）、ビジュアル製品デモなど汎用性が高いですね。

もはや「ウェブサイト」の概念が変わってくる。

[probiex007 @probiex007]

🤯 This is a website, a simple web-based game built with WebGL and Three.js.
Website: https://t.co/G0cDgaWzKe
It's honestly surprising how far web development has come.

[Satya @heysatya_]

The kind of design that gets you funding and downloads.

One of our favourite app concepts we’ve worked on.📱

[Matteo Spada @matteo_spada]

This app is making 500k installs monthly and an MRR of $1M!

citing medical sources (Harvard) → gives authority and trust

visualized goal timeline → concrete and realistic outcome

paywall with bright colors "goals 4.2x faster" → higher subscription rate

[Anvisha @anvisha]

Launching today: add mind-bending shader effects to your designs.

No graphic design experience required.

Click a button or ask Moda's agent, we handle the rest.

Beautiful moving swirls, mesh gradients, dithering, and so much more 🍬

[Sai Satvik @saisatvik_]

Designed this landing page for a AI resume-building platform last month for $4000

[QClay Design @qclay_design]

Tried pushing this exploration toward something more expressive, emotional, and product-focused 🚀

[0xMarioNawfal @RoundtableSpace]

SOMEONE USED CLAUDE FABLE 5 TO BUILD A FREE OPEN SOURCE LIBRARY OF SHADERS FOR WEBSITES, VIDEOS, POSTERS, AND MORE

[0xMarioNawfal @RoundtableSpace]

CLAUDE FABLE 5 CAN NOW DOWNLOAD A YOUTUBE VIDEO AND TURN IT INTO A SCHEDULED SOCIAL POST WITH ZERO HUMAN EDITING

it finds the viral moments, adds captions, reframes to the speaker, and posts it automatically, the entire video editing workflow is gone.

[สมายอาย @sssmileye]

นี่ขอยืนยันอีกเสียงว่า Shopee Affiliate คือเป็นอาชีพเสริมที่เริ่ดมาก 3เดือนสิ้นปีหมดไฟกับการทำงานมาก เลยลองมาศึกษาดู โพสของที่มีในบ้านลงในfbโง่ๆเลย ละโพสไวรัล สรุปได้เดือนละแสน บ้าปะไปอยู่ที่ไหนมา ทำไมพึ่งเคยทำวะ

[Asaad Mahmood - The Small Square @AsaadMahmood5]

Designed this site in Fable 5 with just a few prompts.

And no, this is not a lame post where I attach my actual (not AI) work on an AI post to get views by making fun of a launch.

This is actually Fable 5, as you can clearly see (with some of the elements that scream AI generated).

But man is it doing a great job!

[FHILY👑 @Oluwaphilemon1]

How I build with Claude Fable 5

1. CSS position sticky = how a product follows you down the page

2. Monospace fonts = instantly signal raw and authentic

3. CSS object-fit cover = full bleed photography fills any screen

4. GSAP ScrollTrigger = pins product while background scrolls

position fixed + transform = floating centered label

5. CSS grid + z-index = editorial depth flat sites lack

Save this 🏔️

[Viktor Oddy @viktoroddy]

Claude Mythos is Insane!

❤️‍🔥Just recorded a 12-min tutorial on how to build animated, award-Winning websites with Claude Fable 5!

[Coin Shot ☁️ @CoinSh0t]

This girl alone crashed the stocks of Figma and Photoshop.

Her new technology "touchdesign" made her $420,000

It's already used in Nolan's "Oddysey" movie.

She had the idea, and Claude built it from scratch.

I reverse-engineered this and was shocked by how easy it was to recreate:

The whole system runs on your laptop with just a webcam and two free programs.

MediaPipe reads your webcam and tracks twenty-one points on each hand.

TouchDesigner turns those points into a 3D object that reacts to every move you make.

1. Download TouchDesigner from https://t.co/u5ttgWK7FJ with the free non-commercial license.

2. Grab the MediaPipe plugin from https://t.co/29A1w2XzxL and drag the file into TouchDesigner, no installation needed.

3. Click the hand_tracking node and set numHands to 2 so both your hands get tracked in real time.

4. Open Claude Code in the project folder and paste this prompt:

{ Write me a TouchDesigner Python script that creates a 3D glitchy
cube controlled by both hands using the MediaPipe plugin.

Controls:

- Midpoint between both hands moves the cube around the screen

- Angle between the hands rotates the cube on X and Y axes

- Distance between both hands scales the cube up or down

- Left-hand pinch increases the noise distortion

- Right-hand pinch cycles through colors

- Fast movement triggers a glitch flash

- Both fists freeze the frame for a screenshot

Output one Python DAT script with comments so I can tune each value. }

5. Paste the script into a Python DAT node and wire the MediaPipe outputs into it.

6. Press Play and the cube appears floating in front of you, moving with your hands.

7. Swap the cube for any other object later, and all the hand controls keep working the same way.

The girl performs these sessions live at fashion shows and launches.

She charges five to fifteen thousand dollars per event.

She also sells preset packs on Gumroad for ninety-nine dollars each.

Her webcam cost eighty dollars and her software cost nothing.

The whole effect was built in one weekend with Claude.

[Blaze @browomo]

This Chinese student used to charge $1499 for 15 seconds of studio-grade animation. Today AI does the same thing for $140 right inside his own After Effects.

Now the model sets the keyframes itself, drives the camera and computes the easing curves. He used to crank out every one of those frames by hand, right there in the interface.

His work was always pure taste and craft. He kept the timing, the composition, the easing and every button of After Effects and Cinema 4D in his head. Studio quality rested on exactly these people. On living animators. They knew how to build the concept, the storyboard and the final render by hand.

And the collapse turned out to be insultingly simple. It all came down to a bridge between the model and the core of After Effects. An AE MCP Bridge spun up locally. Through JSX scripts the model writes clean ExtendScript itself. It plans the effects, assembles the compositions and computes the camera shake. And it does this faster than the student can storyboard his own idea.

In this setup there's no animator anymore, no storyboards, no manual keyframing. The whole job is carried by just the model, ExtendScript and a single start_render command.

Ready-made studio time still costs many times more on the market. His old rate reached $1500 per clip and ate up 2 or 3 days at the timeline.

The bridge itself was also written by the model, not him.

Inside there's no pretty shell. What lives there is a bare pipeline with exact steps. The input is a folder of photos and one human sentence. Then in seconds the model writes clean ExtendScript, builds New FX comps, adds camera shake and 3D transitions. And the output is a finished .aep project. The only thing you fix by hand is the text.

And this is where it gets interesting.

The scene no longer needs to be built by hand. A marketer just drops in a folder and tells the model in plain human words, without a single technical term. Something like "make these photos into an aggressive 15-second TikTok edit, red and black."

That's it. Click. The script wrote itself. Built the graph and made the transitions. And in just 1.5 hours it dropped a finished project with no human involved at all.

Let's put the economics together. The picture comes out like this.
// an agency charges $500 to $1500 for 15 seconds of graphics and spends 2 or 3 days on it
// with the model the same clip costs $140 and 1.5 hours
// one freelancer easily handles a full 10 commercial orders at once
// stock templates from Envato and MotionArray turn to dust
// a margin of nearly 100% goes to whoever wraps this bridge into a finished agent

His own words at 3 a.m. sound more honest than any analysis. He admits himself that this thing learned from just 3 of his projects. And it already does motion noticeably better than he does. In the end the student outran his master on the 3rd lesson.

But the most honest part of this whole story is simple. He names his mistake himself. And it's not in the animation at all, it's in the positioning.

He tied his income to manual keyframes and to hours at the timeline. He keeps selling those hours. And that's why he inevitably stays behind. The machine renders faster than he can sketch out a storyboard.

He spells out the right move himself, too.

It's the shift from the role of a pixel generator to the role of a systems architect. Then you're no longer selling your hours. You're managing compute. You wrap that same bridge into an autonomous agent. It calmly runs orders around the clock.

Sadly. This year I've seen plenty of stories about creative professions disappearing. This one is the most honest. $1500 per clip collapsed into $140 of tokens. A living student lost to his own script. A whole task is now stated in human words instead of a hand-built scene. And right away, out loud, comes the admission of a wrong business model.

The barrier to entry into studio-level motion just dropped to the level of "describe the edit in words." And now only one single question remains.

Who will be the first to stop selling their hours at the timeline and start managing someone else's compute?

[Nikiton @Nikitont]

15 GITHUB REPOSITORIES THAT WILL SAVE YOU HUNDREDS OF HOURS.

This time the focus is on AI agents and tools for Claude and Codex.

• AI Website Cloner: give a URL, get a working layout copy you can edit
• career-ops: automates job applications. Drop a list, it writes the cover letters and submits
• Caveman: cuts up to 65% of tokens in Claude Code without losing quality
• Avoid AI Writing : removes AI patterns and triggers from generated text
• agent-skills: ready-made skill set for AI agents from a Google engineer
• claude-howto : visual guide to Claude Code with examples. Perfect for beginners.
• boneyard: auto-generates skeleton loading screens in one prompt
• My Brain Is Full Crew: 8+ AI agents that manage your Obsidian. Auto-tag, sort and link notes.
• SEOMachine: built for long-form SEO articles, not short posts
• FreeMocap: body motion capture through a regular webcam
• SocratiCode: indexes your codebase so any AI assistant understands it. No API keys needed.
• Claude Code Templates: CLI for configuring and monitoring Claude Code in real time
• Awesome Codex Skills: curated skill collection for OpenAI Codex
• VoxCPM: speech generation without a tokenizer. Voice design from scratch and cloning.
• Pixelle-Video: automates short video creation. Generates images, voiceover and edits the clip.

Worth more than most paid AI toolkits combined.

To find these repositories, we take the names and search in the GitHub search.

[Yarchi @undefinedKi]

REAL ESTATE PEOPLE WILL HATE HIM FOR THIS. HE BUILT A CLAUDE AGENT THAT TURNS ANY LISTING INTO A SELLABLE VIDEO ON ITS OWN

Playbook: connect Claude to a video generator, paste a listing, get a cinematic tour of every room, sell it to the agent

But typing the prompt for every listing doesn't scale. He turned it into a skill his Claude runs on its own

Here's how to build the automated version:

1. Connect the video engine once. In Claude, go to Customize, Connectors, Add Custom Connector, name it Higgsfield, and paste the server URL from higgsfield. ai/mcp. Authenticate through your account. No API keys. Now Claude can generate video straight from chat

2. Turn the workflow into a skill. Instead of pasting the same prompt every time, have Claude build a skill. Tell it: "Create a skill called listing-to-video. When I give it a listing URL, scrape the room photos, generate a cinematic clip of each room with Higgsfield, and save them to a folder." Now the whole process is one command, not a wall of text

3. Let the agent run the listing. Hand it a URL and say "run listing-to-video on this." It pulls the photos, fires each room through the video model, and brings the clips back. You wrote the prompt once, inside the skill. You never write it again

4. Stitch and deliver. Drop the clips together into one tour. Send a free sample to the listing's agent, then charge per video or a monthly rate for ongoing listings

5. Scale it with your team. Add a skill that drafts the outreach email and one that builds a simple landing page for the agent. Now one operator runs sourcing, production, and pitching from a single Claude session

The edge isn't generating one video. It's building the skill once so every future listing runs itself

Bookmark this

[Ashik @AshikArtistry]

Everything looks good in motion. Stay moving. 💨