Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

AI video editor in Claude Code: 3 prompts, from an empty folder to a finished reel. SETUP (once) 1. Get the Claude app (claude.ai/download) and open the Code tab, not the browser. 2. Make a folder "AI Editor" with 3 folders inside: raw, references, done. 3. New chat in... Code → Local → Select folder → AI Editor. Needs a paid Claude plan. PROMPT 1: TOOLS You're my video editor. We work in this folder: raw, references, done. Install and verify 5 tools: FFmpeg with libass (on Mac: ffmpeg-full, full path), faster-whisper for word-level subtitles, yt-dlp for references, Remotion, HyperFrames (npm package hyperframes). Check what's installed first. Official installs only (Homebrew, winget, npm/pip), Node.js 22+. Python packages go in a .venv here, Python 3.14 or older: pip install faster-whisper "av<19". Test each for real: burn a subtitle into a test clip, transcribe 5 seconds of audio. Show a table: tool, version, works or not. PROMPT 2: YOUR TASTE Here are reels whose editing I like: [link 1] [link 2] [link 3]. My format: [talking to camera / voiceover]. Download them with yt-dlp into references (for study only). Study each one frame by frame and by ear: pacing, shot length, pause cuts, subtitles, colors, graphics, transitions, music, the first 3 seconds. Take inspiration, but don't copy: no shots, text, logos or brand elements. Write my own style to style.md with exact values. Show it and ask what to change. PROMPT 3: EDIT My new recording is in raw. Edit it using style.md: 1. Transcribe with Whisper, word timestamps. Ask me about doubtful words before burning subtitles. 2. Cut pauses, slips and bad takes, keep the last good take. Check cut points by audio level. Keep pauses that serve the meaning, breaths and word endings. Don't speed up my voice. 3. Word-by-word subtitles in my style. 4. If references/music.mp3 exists, put it quietly under the voice. 5. Graphics only where they help. 6. Export 1080×1920 to done as draft-v1.mp4 plus edit-notes.md. Check the file yourself, then show me. I'll give edits in plain words; save each as draft-v2, draft-v3. No code words. Follow for my own editing style, coming next.show more

Arti Miller

7,858 subscribers

723,800 views • 8 days ago •via TikTok

Education Arts Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Motion graphics used to take me an hour per scene (or more) Now I'm creating them in 3-5 minutes — without After Effects, without Premiere, without writing a single line of code. I'm a former LA filmmaker (still PGA). My entire video stack is now 3 tools: Claude Code + HyperFrames (from HeyGen) + Descript Here's how the workflow runs: 1. Drop my brand style guide into a folder (colors, fonts, components Claude reuses forever) 2. Paste a line from my Descript transcript into Claude Code: "Create a scene for this part of the video" 3. Claude Code reads my brand, generates the HyperFrames scene, renders it to MP4 4. Drag the MP4 into Descript, drop it on the matching line of the transcript 5. Done. 3-5 minutes per scene. The unlock is the brand style guide. Once Claude has your colors, fonts, and reusable components, every scene inherits your look automatically. Bonus — you can take a Claude Design export (.zip), hand it to Claude Code, and it'll convert the HTML into a HyperFrames video you can render and edit. I built a GitHub repo with all the skills pre-loaded. Non-developers can clone it, open Claude Code, and be rendering scenes in under 10 minutes. Full walkthrough (20 min video) below Repo (skills + brand template + music-from-transcript plugin): Bookmark this if you make videos. The brand-style-guide pattern alone will save you hours every week. If you're a creator or operator using AI for content — share this with someone who's still living in After Effects.

JJ Englert

103,908 views • 6 months ago

Can your agent finally edit your videos? Yup, it's a repo called video use, by Browser Use. You put the raw footage in a folder. Then Claude Code or Codex edits it. It removes the filler words, cuts the dead space, adds the subtitles, and renders the file. Then it checks its own cut points. Most creators use the same small set of edits in every video. That style stays in your head and in your hands. This repo puts it in a file that an agent can read. People ask for a full YouTube channel on day one. They see bad results, and they stop. Start with one format instead. - One founder loom into a 60 second launch video - 3 clips out of one podcast recording Then make the format a system. Then you operate the system, or you sell it. - Real estate agents need listing videos - Coaches need clips - Agencies need ad variations Manual installation with git clone is possible. For 95% of people, the agent is faster. Here's how i intalled it: - Gave the setup prompt to Claude Code with shell access - Let it read install.md - Installed the repo - Connected FFmpeg - Registered the skill

Can your agent finally edit your videos? Yup, it's a repo called video use, by Browser Use. You put the raw footage in a folder. Then Claude Code or Codex edits it. It removes the filler words, cuts the dead space, adds the subtitles, and renders the file. Then it checks its own cut points. Most creators use the same small set of edits in every video. That style stays in your head and in your hands. This repo puts it in a file that an agent can read. People ask for a full YouTube channel on day one. They see bad results, and they stop. Start with one format instead. - One founder loom into a 60 second launch video - 3 clips out of one podcast recording Then make the format a system. Then you operate the system, or you sell it. - Real estate agents need listing videos - Coaches need clips - Agencies need ad variations Manual installation with git clone is possible. For 95% of people, the agent is faster. Here's how i intalled it: - Gave the setup prompt to Claude Code with shell access - Let it read install.md - Installed the repo - Connected FFmpeg - Registered the skill

The Startup Ideas Podcast (SIP) 🧃

15,605 views • 2 months ago

OFF THE PAGE (ページの外へ) A mangaka spills ink on her last page at 3:47 a.m. The monster she crossed out crawls off the paper, and her hero tears out of his panel to stop it. 60 seconds, made start to finish in #PixAIStudio by PixAI . Here's exactly how: 1. Characters first. Three Tsubaki.3 turnaround sheets: the mangaka, the swordsman and an ink creature, with the Ultimate anime style and Beautiful colors V2 LoRAs for a TV-anime look. 2. The manga page is the storyboard. One 6-panel Tsubaki.3 page with the sheets connected as references. It plans the shots and it's the prop on her desk. 3. Key frames. PixAI Edit Pro v2.0 with the room and character sheets as references, so the apartment stays the same in every shot. 4. Video. Four 15s clips in V4.0 at 1080p, each key frame connected straight into a Video node. Every prompt is a cut list: shot, who stands where, action, lighting. 5. Music. Two 30s Audio nodes: piano, rain and taiko drums. 6. Edit. Sequence Editor: drop the clips on the timeline, trim, Export MP4. What didn't work first time: - The trigger word "textured hair" gave my ink monster a horse's mane. Removing it fixed it. - My first manga page came out as ink blobs. Asking for "3 rows of 2 panels" fixed the layout. - Without references, the room changed between shots. Edit Pro fixed that. - The Audio node failed with my long, timed music prompt. A short, simple prompt at 30s worked. Tips: test your hardest shot first, and connect references instead of re-describing characters. Shorter prompts, steadier faces.

OFF THE PAGE (ページの外へ) A mangaka spills ink on her last page at 3:47 a.m. The monster she crossed out crawls off the paper, and her hero tears out of his panel to stop it. 60 seconds, made start to finish in #PixAIStudio by PixAI . Here's exactly how: 1. Characters first. Three Tsubaki.3 turnaround sheets: the mangaka, the swordsman and an ink creature, with the Ultimate anime style and Beautiful colors V2 LoRAs for a TV-anime look. 2. The manga page is the storyboard. One 6-panel Tsubaki.3 page with the sheets connected as references. It plans the shots and it's the prop on her desk. 3. Key frames. PixAI Edit Pro v2.0 with the room and character sheets as references, so the apartment stays the same in every shot. 4. Video. Four 15s clips in V4.0 at 1080p, each key frame connected straight into a Video node. Every prompt is a cut list: shot, who stands where, action, lighting. 5. Music. Two 30s Audio nodes: piano, rain and taiko drums. 6. Edit. Sequence Editor: drop the clips on the timeline, trim, Export MP4. What didn't work first time: - The trigger word "textured hair" gave my ink monster a horse's mane. Removing it fixed it. - My first manga page came out as ink blobs. Asking for "3 rows of 2 panels" fixed the layout. - Without references, the room changed between shots. Edit Pro fixed that. - The Audio node failed with my long, timed music prompt. A short, simple prompt at 30s worked. Tips: test your hardest shot first, and connect references instead of re-describing characters. Shorter prompts, steadier faces.

Iris

22,637 views • 5 days ago

Claude Code is now a full video editor! Our new plugin opens a full built-in NLE right inside the Claude desktop app, where you and your agent can edit together, effortlessly. If you're already paying for Claude, you get this one for FREE. It's available now. Start using it in 1 minute, by pasting this prompt into any Claude Code chat: "Read to install and use the ChatCut plugin"

Claude Code is now a full video editor! Our new plugin opens a full built-in NLE right inside the Claude desktop app, where you and your agent can edit together, effortlessly. If you're already paying for Claude, you get this one for FREE. It's available now. Start using it in 1 minute, by pasting this prompt into any Claude Code chat: "Read to install and use the ChatCut plugin"

ChatCut

227,831 views • 3 months ago

Claude Code can now make full videos from your terminal.. Not slideshows. Not text on screen.. Actual motion graphics with animations, transitions, custom photos, and background music. [ SHARED A TUTORIAL BELOW EDITED WITH THIS SETUP IN JUST 5mins ] ▫️Here's the setup: Claude Code + Remotion Remotion is a React based framework that renders video programmatically. You describe what you want in plain English, Claude writes the React components and Remotion renders it into a real MP4. What you can actually do with this: > Generate 9:16 vertical videos for TikTok / Reels / Shorts > Add animated text with viral hooks and safe zones > Pull live web screenshots directly into your scenes using Chrome MCP > Fact-check your content in real time with Perplexity MCP > Drop in your own photos and background music > Edit existing talking-head footage cut bloopers, add captions > Schedule posts to your socials straight from the terminal ▫️How to set it up (takes 5 minutes) : > Make sure you have Node.js installed ( node -v to check ) > Create a new Remotion project: npx create-video@latest Pick the Blank template, enable TailwindCSS, and install the Skills package when prompted. > Install dependencies: cd my-video npm install > Start the preview server: npm run dev > Open Claude Code in the same project folder: cd my-video claude That's it. You can now prompt videos in plain English. If you already have a Remotion project, just add the skill directly: npx skills add remotion-dev/skills This drops a SKILL.md into your project that gives Claude expert knowledge of Remotion.. animations, compositions, captions, assets, 3D content everything. Example prompt you can steal: "Create a 30-second 9:16 vertical video about the top 3 AI tools this week. Use animated text with a hook in the first 2 seconds. Add smooth transitions between scenes. Keep text in the safe zone for TikTok. Use a dark tech aesthetic with blue accent colors." Claude writes all the React code, renders a preview, you tweak with natural language, and export when ready. The crazy part is this whole pipeline is local, free (minus your Claude sub), and you never open a video editor. imo this kills CapCut for anyone making info-style content. You describe the video in English and get back a rendered MP4. try it now.

Claude Code can now make full videos from your terminal.. Not slideshows. Not text on screen.. Actual motion graphics with animations, transitions, custom photos, and background music. [ SHARED A TUTORIAL BELOW EDITED WITH THIS SETUP IN JUST 5mins ] ▫️Here's the setup: Claude Code + Remotion Remotion is a React based framework that renders video programmatically. You describe what you want in plain English, Claude writes the React components and Remotion renders it into a real MP4. What you can actually do with this: > Generate 9:16 vertical videos for TikTok / Reels / Shorts > Add animated text with viral hooks and safe zones > Pull live web screenshots directly into your scenes using Chrome MCP > Fact-check your content in real time with Perplexity MCP > Drop in your own photos and background music > Edit existing talking-head footage cut bloopers, add captions > Schedule posts to your socials straight from the terminal ▫️How to set it up (takes 5 minutes) : > Make sure you have Node.js installed ( node -v to check ) > Create a new Remotion project: npx create-video@latest Pick the Blank template, enable TailwindCSS, and install the Skills package when prompted. > Install dependencies: cd my-video npm install > Start the preview server: npm run dev > Open Claude Code in the same project folder: cd my-video claude That's it. You can now prompt videos in plain English. If you already have a Remotion project, just add the skill directly: npx skills add remotion-dev/skills This drops a SKILL.md into your project that gives Claude expert knowledge of Remotion.. animations, compositions, captions, assets, 3D content everything. Example prompt you can steal: "Create a 30-second 9:16 vertical video about the top 3 AI tools this week. Use animated text with a hook in the first 2 seconds. Add smooth transitions between scenes. Keep text in the safe zone for TikTok. Use a dark tech aesthetic with blue accent colors." Claude writes all the React code, renders a preview, you tweak with natural language, and export when ready. The crazy part is this whole pipeline is local, free (minus your Claude sub), and you never open a video editor. imo this kills CapCut for anyone making info-style content. You describe the video in English and get back a rendered MP4. try it now.

Axel Bitblaze 🪓

31,862 views • 7 months ago

Introducing Reel Farmer Paste a YouTube URL, get TikTok/Reel with animated captions in one command. Fully local. > How it works: 1. yt-dlp downloads the video 2. LLM reads the transcript and picks the most clip-worthy moments 3. FFmpeg cuts each clip, silence removed automatically 4. Whisper.cpp runs locally for word-level timestamps 5. Remotion renders animated captions 6. Combines everything into 1080x1920 reels Every stage checkpoints to SQLite. In case of a crash, it'll pick up exactly where it left off. MiniMax (official) M2.7 wrote the entire code.

Introducing Reel Farmer Paste a YouTube URL, get TikTok/Reel with animated captions in one command. Fully local. > How it works: 1. yt-dlp downloads the video 2. LLM reads the transcript and picks the most clip-worthy moments 3. FFmpeg cuts each clip, silence removed automatically 4. Whisper.cpp runs locally for word-level timestamps 5. Remotion renders animated captions 6. Combines everything into 1080x1920 reels Every stage checkpoints to SQLite. In case of a crash, it'll pick up exactly where it left off. MiniMax (official) M2.7 wrote the entire code.

shydev

148,826 views • 7 months ago

It is becoming f*cking dangerous: Claude Code has kept its own memory folder since February and you've probably never opened it 1. Open any project in Claude Code and type /memory 2. Check the auto memory toggle, it is on by default in local sessions 3. Pick the option to open the auto memory folder 4. Read MEMORY.md first, it is the index, one line per memory 5. Look at the files: one memory each, typed user, feedback, project or reference 6. Tell Claude "remember: always use pnpm, not npm" and watch it save a new memory 7. Edit or delete any wrong memory, it is plain markdown Only the first 200 lines or 25KB of MEMORY.md load each session, Claude opens the topic files when needed Ask Claude to remember what matters once, it stays until you or Claude edit or delete it Full guide in the article below

It is becoming f*cking dangerous: Claude Code has kept its own memory folder since February and you've probably never opened it 1. Open any project in Claude Code and type /memory 2. Check the auto memory toggle, it is on by default in local sessions 3. Pick the option to open the auto memory folder 4. Read MEMORY.md first, it is the index, one line per memory 5. Look at the files: one memory each, typed user, feedback, project or reference 6. Tell Claude "remember: always use pnpm, not npm" and watch it save a new memory 7. Edit or delete any wrong memory, it is plain markdown Only the first 200 lines or 25KB of MEMORY.md load each session, Claude opens the topic files when needed Ask Claude to remember what matters once, it stays until you or Claude edit or delete it Full guide in the article below

plusxbt

19,632 views • 12 days ago

Claude Opus 5.5 - Manga Edit Prompt Share Saw a Makima edit while scrolling and thought I'd give it a try with Opus using one of my Makima videos. It created this entirely with code, using just the video from the quoted post and a music file as references. Definitely gonna try a few more edits like this! Prompt in the replies. 👇

Claude Opus 5.5 - Manga Edit Prompt Share Saw a Makima edit while scrolling and thought I'd give it a try with Opus using one of my Makima videos. It created this entirely with code, using just the video from the quoted post and a music file as references. Definitely gonna try a few more edits like this! Prompt in the replies. 👇

Kōda

26,143 views • 18 days ago

I gave Opus 5.5, GPT-6 Astra and Fable 5.1 the same raw video, the same brief and my real YouTube editing process (the one I use for my own videos): 1. Give each model the same footage and reference. 2. Ask to cut the pauses and turn your words into visuals. 3. Review the first 10 seconds, then fix it round by round. Which edit came out better? You can get the guide here:

I gave Opus 5.5, GPT-6 Astra and Fable 5.1 the same raw video, the same brief and my real YouTube editing process (the one I use for my own videos): 1. Give each model the same footage and reference. 2. Ask to cut the pauses and turn your words into visuals. 3. Review the first 10 seconds, then fix it round by round. Which edit came out better? You can get the guide here:

Charlie Hills

93,955 views • 1 month ago

There's a GitHub tool that lets AI agents edit videos like a professional editor. Color grading, trimming, transitions, subtitles, animation overlays all of it. Give the repo to Claude Code and it starts editing in real time. Free. Local. No API keys. One prompt to set up. Video editors are cooked. Credits: m0h Github:

There's a GitHub tool that lets AI agents edit videos like a professional editor. Color grading, trimming, transitions, subtitles, animation overlays all of it. Give the repo to Claude Code and it starts editing in real time. Free. Local. No API keys. One prompt to set up. Video editors are cooked. Credits: m0h Github:

0xMarioNawfal

66,397 views • 4 months ago

How to run GLM-5.2 (Or Minimax or Qwen or Deepseek or Kimi) in Claude Code No JSON editing. No vim 😮‍💨 1. Install Claude Code npm install -g Paul Jankura-ai/claude-code 2. Grab a Fireworks API key → 3. Install FireConnect + drop in your key (one line): curl -fsSL | FIREWORKS_API_KEY="fw_..." bash 4. Flip it on: fireconnect claude on 5. Switch to GLM-5.2 and bam! Swap models with: fireconnect claude model select Docs to do the same for Pi, Opencode, and Codex:

How to run GLM-5.2 (Or Minimax or Qwen or Deepseek or Kimi) in Claude Code No JSON editing. No vim 😮‍💨 1. Install Claude Code npm install -g Paul Jankura-ai/claude-code 2. Grab a Fireworks API key → 3. Install FireConnect + drop in your key (one line): curl -fsSL | FIREWORKS_API_KEY="fw_..." bash 4. Flip it on: fireconnect claude on 5. Switch to GLM-5.2 and bam! Swap models with: fireconnect claude model select Docs to do the same for Pi, Opencode, and Codex:

Sinan Ozdemir

149,913 views • 4 months ago

I'm not an editor. Claude Opus 5.5 edited my entire YouTube video. The stack: 1. Tella for the raw clips 2. DaVinci Resolve 21.1, driven over MCP 3. Hyperframes for B-roll and animation 4. Claude Opus 5.5 as the editor The prompt and how the pieces fit together are in the below video and article. 👇

I'm not an editor. Claude Opus 5.5 edited my entire YouTube video. The stack: 1. Tella for the raw clips 2. DaVinci Resolve 21.1, driven over MCP 3. Hyperframes for B-roll and animation 4. Claude Opus 5.5 as the editor The prompt and how the pieces fit together are in the below video and article. 👇

Nick Spisak

10,375 views • 1 month ago

last week I talked about claude code being my video editor, but it was only a small 30 second clip - here is my FULL workflow on how i have claude code edit ~30 minutes of raw video into a tight 10 minutes it uses STT for captions, then uses remotion to add the captions & cut!

last week I talked about claude code being my video editor, but it was only a small 30 second clip - here is my FULL workflow on how i have claude code edit ~30 minutes of raw video into a tight 10 minutes it uses STT for captions, then uses remotion to add the captions & cut!

BOOTOSHI 👑

14,107 views • 9 months ago

Video editing has nowadays become a conversation.. you drop raw footage in a folder, tell Claude Code what you want, and get final.mp4 back. its free, open source, 17,000 stars on Github.. it's called video-use. and the reason it's different from every "AI video editor" is how it actually works under the hood.. it doesn't guess. it transcribes every word you said first, then edits off the actual transcript. what it does once your footage is in the folder: - cuts every umm, uh, and false start automatically - kills the dead space between takes - color grades each segment.. warm cinematic, neutral punch, or a custom ffmpeg chain - builds the cut with editor sub-agents, runs the animations in parallel - renders a preview, grades its own work, and iterates until it's clean and yeah.. it asks before it cuts. it shows you the plan in plain english and waits for your yes. it's not a black box that spits out a mystery edit.. you stay the director, it does the labor. ▫️ how to start install the skill in claude code (repo below) drop your raw clips in a folder tell it "cut the filler, tighten the pauses, warm cinematic grade" and let it work And yes, it's a tool, not a taste replacement. it'll hand you a clean, tight cut, but the creative calls, the hook, the pacing, the story.. that's still you. it does the 80% that's grunt work so you spend your time on the 20% that matters. editing stopped being a skill you grind for years. it's a prompt now. repo:

Video editing has nowadays become a conversation.. you drop raw footage in a folder, tell Claude Code what you want, and get final.mp4 back. its free, open source, 17,000 stars on Github.. it's called video-use. and the reason it's different from every "AI video editor" is how it actually works under the hood.. it doesn't guess. it transcribes every word you said first, then edits off the actual transcript. what it does once your footage is in the folder: - cuts every umm, uh, and false start automatically - kills the dead space between takes - color grades each segment.. warm cinematic, neutral punch, or a custom ffmpeg chain - builds the cut with editor sub-agents, runs the animations in parallel - renders a preview, grades its own work, and iterates until it's clean and yeah.. it asks before it cuts. it shows you the plan in plain english and waits for your yes. it's not a black box that spits out a mystery edit.. you stay the director, it does the labor. ▫️ how to start install the skill in claude code (repo below) drop your raw clips in a folder tell it "cut the filler, tighten the pauses, warm cinematic grade" and let it work And yes, it's a tool, not a taste replacement. it'll hand you a clean, tight cut, but the creative calls, the hook, the pacing, the story.. that's still you. it does the 80% that's grunt work so you spend your time on the 20% that matters. editing stopped being a skill you grind for years. it's a prompt now. repo:

Axel Bitblaze 🪓

117,388 views • 3 months ago

Here's an MS Word feature that nobody is talking about and that just blew my mind. Open an blank document in MS Word and click on the "Dicatate" button. Select "Transcribe." Next upload an audio file you want to transcribe. I uploaded recording of one my webinars. It's 3 hours and 17 minutes long. And MS Word will transcribe the whole thing for you in a few minutes. Once it's done transcribing, Word gives you four options: Just text, With speakers, With timestamps, With speakers and timestamps. Choose the one according to your preference and you will have the whole transcript.

Here's an MS Word feature that nobody is talking about and that just blew my mind. Open an blank document in MS Word and click on the "Dicatate" button. Select "Transcribe." Next upload an audio file you want to transcribe. I uploaded recording of one my webinars. It's 3 hours and 17 minutes long. And MS Word will transcribe the whole thing for you in a few minutes. Once it's done transcribing, Word gives you four options: Just text, With speakers, With timestamps, With speakers and timestamps. Choose the one according to your preference and you will have the whole transcript.

Mushtaq Bilal, PhD

343,942 views • 3 years ago

i just made this stickman animation in full code with claude opus 5.5 took me 20 mins and spent $0 on videos or images promised you guys id give a free claude skill to re create any code format so here it is hand this prompt to your claude code and thank me later ↓ Ask me for: my video idea (one line is fine), the setting, the length (60-90 s) and the narrator (an ElevenLabs voice, or "free" for a local text-to-speech). Everything is built in code: no AI images, no AI video. Make it MY story, not a copy of an example. The style is a second-person survival thriller with stickmen, built as a series: - "Imagine you…" hook with a time stamp in the first 3 s - a HUD with the DAY number and a clock that moves forward through the episode - 3-4 RULES, each one paid off on screen - a threat that escalates and is mostly shown, not told - a twist that makes one rule impossible - a two-choice ending ("DO YOU OPEN IT?" style, a green and a red option, "COMMENT BELOW") and "FOLLOW FOR DAY 2" Around 20-30 voice lines, 60-90 s, a new shot every 1.5-3 s. Write a beat sheet and the script first and show me both before building anything. - One line = one visual. Rules are short commands. - Open a new question every ~10 s and only answer it with a bigger one. - Put the viewer in POV for the scary parts: what you see through a door, a window, a phone, a peephole. - The ending is a choice where both options look bad. - Write numbers twice: the caption form ("11:58 PM") and the spoken form ("eleven fifty-eight P M") for the text-to-speech. - Rotate settings between episodes (apartment, subway, school at night, hospital, cabin, plane, office, motel) and fit the threat to the setting. Pick a palette per story, but keep the language: - real 3D low-poly sets with real lights and shadows; the mood comes from light (one warm practical, one cold fill, one strong accent such as a door gap, a screen or a siren) - the stickman is flat and 2D-looking: a light body with a thick dark outline, round strokes, dot eyes, no mouth; give the main character one mark (a scar, a cap, a colour) so doubles and callbacks read - the overlay: a monospace HUD top-left, chunky captions (1-3 words, key words in an accent colour), a tilted rule sticker with a big condensed title, phone screens drawn in 2D - banned: AI frames, glossy 3D characters, emojis, music with vocals, long captions A library to adapt to the story (never use all of them): - the opener: a camera move that starts tight on you and pulls out to show where you are - POV phone in your hands (2D phone UI, two stick thumbs) - a macro on the one detail that matters - outside chaos from above (balcony, bridge, window): physics vehicles, a running crowd, something collapsing or burning - third-person action in the room: walking with planted feet, reaching, hesitating, barricading - POV interaction: throwing, holding a torch, opening something - a close-up of your face for the reaction - a low angle on a door or vent: light leaking in, shadows moving - knocks, shakes, things falling - a fisheye peephole, camera or window view of the threat stepping closer - the choice screen and the end card What each tool is for. Install whatever is missing, and ask me before anything paid: - three . js: the 3D engine; it runs inside a web page. Download its module file locally so renders don't depend on the internet. Use its post-processing passes for bloom, plus one custom lens shader for vignette, grain, fisheye and colour fringing. - cannon-es: a JavaScript physics engine for vehicles, falling debris and collapses. - Playwright (Python): opens the page in a real browser (Edge or Chrome with the GPU on) and captures every frame as an image. - ffmpeg: the command-line video and audio tool. It joins the frames into an MP4, cuts and filters the audio (highpass, lowpass and echo for voices behind doors), mixes, normalises loudness and makes the phone-size copy. - ElevenLabs, or Kokoro (free local text-to-speech), for the narrator. - faster-whisper: free local speech-to-text with word timestamps. Use it to cut the voice cleanly and to time every caption. - numpy and Pillow: to check audio levels and make contact sheets of stills. - Sound: free downloaded sound effects from Mixkit (never synthesized) and a dark cinematic drone or tension bed with no vocals. 1. One HTML page (film . html) with the 3D renderer at 1080x1920 drawn into a 2D canvas that also holds the overlay. A function window . seek (t) draws frame t and must be a pure function of time; every cut, sound and move is keyed to the word times. 2. Stickman. Draw each figure with 2D canvas strokes onto its own texture and put it on a plane that always faces the camera, with an alpha test so it casts a real shadow and slight emissive so it reads in the dark. Pose with joint angles, and walk with 2-bone IK so the planted foot never slides. 3. Physics and cloth. Simulate once at load with a fixed timestep, record the positions, then sample them by time, so every render is identical. Cloth (sheets, curtains, flags) is a verlet particle grid with wall and floor collisions and pinned points you can release. 4. Voice. If you generate one long take, never cut it at the API's character timestamps: they drift and slice words. Transcribe the take with faster-whisper word timestamps, match it to the script, cut only in real silence between lines, and only add pause where the natural one is too short. 5. Captions. Re-transcribe each final line, normalise numbers ("6:42" = "6 42", "AM" = "a m"), align the caption words to what was heard, and snap each word to its real start in the audio. Check that every caption has voice under it within ~120 ms. 6. Render. Playwright steps window . seek through every frame at 30 fps, and ffmpeg encodes a high-quality master, then a phone copy under 30 MB. 7. Mix with ffmpeg: the voice on top, the music about 12-14 dB under it, one sound effect per event, a dip before the big moment, loudness at -14 LUFS, then a scan for dropouts. - is narrow: a normal lens sees very little sideways, so use wider lenses indoors and keep the subject close to the aim point. - Glass behind a solid frame is invisible: build mirrors and windows as a frame around a real hole. - Dust spread everywhere reads as stars, so keep it inside light beams. - Night shots go black after compression, so give each one a cold fill light. - Fog swallows distant things. - Never judge a reference or your own render from one frame every few seconds; check motion at the full frame rate. - Check cuts, captions and audio joints with numbers before every full render. Ask me for the idea, show me the beat sheet and script, then 6 stills of the key shots before rendering the full video.

i just made this stickman animation in full code with claude opus 5.5 took me 20 mins and spent $0 on videos or images promised you guys id give a free claude skill to re create any code format so here it is hand this prompt to your claude code and thank me later ↓ Ask me for: my video idea (one line is fine), the setting, the length (60-90 s) and the narrator (an ElevenLabs voice, or "free" for a local text-to-speech). Everything is built in code: no AI images, no AI video. Make it MY story, not a copy of an example. The style is a second-person survival thriller with stickmen, built as a series: - "Imagine you…" hook with a time stamp in the first 3 s - a HUD with the DAY number and a clock that moves forward through the episode - 3-4 RULES, each one paid off on screen - a threat that escalates and is mostly shown, not told - a twist that makes one rule impossible - a two-choice ending ("DO YOU OPEN IT?" style, a green and a red option, "COMMENT BELOW") and "FOLLOW FOR DAY 2" Around 20-30 voice lines, 60-90 s, a new shot every 1.5-3 s. Write a beat sheet and the script first and show me both before building anything. - One line = one visual. Rules are short commands. - Open a new question every ~10 s and only answer it with a bigger one. - Put the viewer in POV for the scary parts: what you see through a door, a window, a phone, a peephole. - The ending is a choice where both options look bad. - Write numbers twice: the caption form ("11:58 PM") and the spoken form ("eleven fifty-eight P M") for the text-to-speech. - Rotate settings between episodes (apartment, subway, school at night, hospital, cabin, plane, office, motel) and fit the threat to the setting. Pick a palette per story, but keep the language: - real 3D low-poly sets with real lights and shadows; the mood comes from light (one warm practical, one cold fill, one strong accent such as a door gap, a screen or a siren) - the stickman is flat and 2D-looking: a light body with a thick dark outline, round strokes, dot eyes, no mouth; give the main character one mark (a scar, a cap, a colour) so doubles and callbacks read - the overlay: a monospace HUD top-left, chunky captions (1-3 words, key words in an accent colour), a tilted rule sticker with a big condensed title, phone screens drawn in 2D - banned: AI frames, glossy 3D characters, emojis, music with vocals, long captions A library to adapt to the story (never use all of them): - the opener: a camera move that starts tight on you and pulls out to show where you are - POV phone in your hands (2D phone UI, two stick thumbs) - a macro on the one detail that matters - outside chaos from above (balcony, bridge, window): physics vehicles, a running crowd, something collapsing or burning - third-person action in the room: walking with planted feet, reaching, hesitating, barricading - POV interaction: throwing, holding a torch, opening something - a close-up of your face for the reaction - a low angle on a door or vent: light leaking in, shadows moving - knocks, shakes, things falling - a fisheye peephole, camera or window view of the threat stepping closer - the choice screen and the end card What each tool is for. Install whatever is missing, and ask me before anything paid: - three . js: the 3D engine; it runs inside a web page. Download its module file locally so renders don't depend on the internet. Use its post-processing passes for bloom, plus one custom lens shader for vignette, grain, fisheye and colour fringing. - cannon-es: a JavaScript physics engine for vehicles, falling debris and collapses. - Playwright (Python): opens the page in a real browser (Edge or Chrome with the GPU on) and captures every frame as an image. - ffmpeg: the command-line video and audio tool. It joins the frames into an MP4, cuts and filters the audio (highpass, lowpass and echo for voices behind doors), mixes, normalises loudness and makes the phone-size copy. - ElevenLabs, or Kokoro (free local text-to-speech), for the narrator. - faster-whisper: free local speech-to-text with word timestamps. Use it to cut the voice cleanly and to time every caption. - numpy and Pillow: to check audio levels and make contact sheets of stills. - Sound: free downloaded sound effects from Mixkit (never synthesized) and a dark cinematic drone or tension bed with no vocals. 1. One HTML page (film . html) with the 3D renderer at 1080x1920 drawn into a 2D canvas that also holds the overlay. A function window . seek (t) draws frame t and must be a pure function of time; every cut, sound and move is keyed to the word times. 2. Stickman. Draw each figure with 2D canvas strokes onto its own texture and put it on a plane that always faces the camera, with an alpha test so it casts a real shadow and slight emissive so it reads in the dark. Pose with joint angles, and walk with 2-bone IK so the planted foot never slides. 3. Physics and cloth. Simulate once at load with a fixed timestep, record the positions, then sample them by time, so every render is identical. Cloth (sheets, curtains, flags) is a verlet particle grid with wall and floor collisions and pinned points you can release. 4. Voice. If you generate one long take, never cut it at the API's character timestamps: they drift and slice words. Transcribe the take with faster-whisper word timestamps, match it to the script, cut only in real silence between lines, and only add pause where the natural one is too short. 5. Captions. Re-transcribe each final line, normalise numbers ("6:42" = "6 42", "AM" = "a m"), align the caption words to what was heard, and snap each word to its real start in the audio. Check that every caption has voice under it within ~120 ms. 6. Render. Playwright steps window . seek through every frame at 30 fps, and ffmpeg encodes a high-quality master, then a phone copy under 30 MB. 7. Mix with ffmpeg: the voice on top, the music about 12-14 dB under it, one sound effect per event, a dip before the big moment, loudness at -14 LUFS, then a scan for dropouts. - is narrow: a normal lens sees very little sideways, so use wider lenses indoors and keep the subject close to the aim point. - Glass behind a solid frame is invisible: build mirrors and windows as a frame around a real hole. - Dust spread everywhere reads as stars, so keep it inside light beams. - Night shots go black after compression, so give each one a cold fill light. - Fog swallows distant things. - Never judge a reference or your own render from one frame every few seconds; check motion at the full frame rate. - Check cuts, captions and audio joints with numbers before every full render. Ask me for the idea, show me the beat sheet and script, then 6 stills of the key shots before rendering the full video.

Finn

64,458 views • 13 days ago

$ANTHROPIC JUST QUIETLY SHIPPED A FEATURE THAT LETS CLAUDE SPAWN A WHOLE TEAM OF AGENTS THAT MESSAGE EACH OTHER AND REVIEW EACH OTHER'S WORK. It's a Claude Code feature called agent teams. The team lead spawns multiple agents that share a task list and message each other directly, not subagents reporting back, actual peers. In the demo a QA agent caught three bugs, sent the work back to the front-end and back-end devs, they fixed it, app shipped in one pass. How to run it: 1. Enable it. Needs Claude Code v2.1.32+. Add to settings.json: "env": { "CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS": "1" }. Or paste that to Claude and say "add this to my settings." Restart. 2. Prompt in plain English. Start with a goal (agents wake with zero context), then "create a team of 3 using Sonnet," describe each role, its deliverable, and who it messages when done. 3. The rules: each agent owns its own files, define exact outputs, name who talks to who, keep it to 3-5 agents. Use it for complex work with separate parts running in parallel. Skip it for simple or sequential tasks, teams cost 3-4x the tokens. Bookmark this.$

ANTHROPIC JUST QUIETLY SHIPPED A FEATURE THAT LETS CLAUDE SPAWN A WHOLE TEAM OF AGENTS THAT MESSAGE EACH OTHER AND REVIEW EACH OTHER'S WORK. It's a Claude Code feature called agent teams. The team lead spawns multiple agents that share a task list and message each other directly, not subagents reporting back, actual peers. In the demo a QA agent caught three bugs, sent the work back to the front-end and back-end devs, they fixed it, app shipped in one pass. How to run it: 1. Enable it. Needs Claude Code v2.1.32+. Add to settings.json: "env": { "CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS": "1" }. Or paste that to Claude and say "add this to my settings." Restart. 2. Prompt in plain English. Start with a goal (agents wake with zero context), then "create a team of 3 using Sonnet," describe each role, its deliverable, and who it messages when done. 3. The rules: each agent owns its own files, define exact outputs, name who talks to who, keep it to 3-5 agents. Use it for complex work with separate parts running in parallel. Skip it for simple or sequential tasks, teams cost 3-4x the tokens. Bookmark this.

Yarchi

463,029 views • 4 months ago

This video was edited with 1 prompt with Opus 5.5. I uploaded a single video of myself (with many mistakes). Opus went on to: 1. Analyze the transcript 2. Search Twitter for the relevant B-roll (Claude Code literally calls GrokBot for this step) 3. Scrape the Twitter video assets 4. Cut the video based on mistakes and word gaps 5. Build out the Blender scene from a template 6. Build the UI and components of each shot (my Blender scene has 4 shots: a computer, phone, table, and board) 7. It took screenshots of every section, analyzed and fixed them 8. Synced every shot, zoom, and animation to the exact words I say 9. Animated a robot arm placing real logos on the table right when I mention them 10. Drew the whiteboard ending with logos and graphics instead of text 11. Pulled a cutaway from the creator's own video, with his voice 12. Rendered all 3,487 frames on RunPod cloud GPUs in under 20 minutes, for less than $1 13. Put me in the bottom third only when I'm talking, so the B-roll stays full screen 14. Added word-by-word captions, sound effects, and music 15. Uploaded it to my Google Drive, ready to post Will be making a full youtube video this week explaining it. I want to learn a little more about how cloud rendering works before making video, as well as figuring out what the best provider is. (If you are in cloud rendering space, hit me up) My goal of the youtube video will not be to make it a copyable skill, rather teach people the foundations so they can be creative and make their own mini-world by which they can make videos from, while understanding what's happening under the hood. Shout out Charlie Chapman Max Blade

This video was edited with 1 prompt with Opus 5.5. I uploaded a single video of myself (with many mistakes). Opus went on to: 1. Analyze the transcript 2. Search Twitter for the relevant B-roll (Claude Code literally calls GrokBot for this step) 3. Scrape the Twitter video assets 4. Cut the video based on mistakes and word gaps 5. Build out the Blender scene from a template 6. Build the UI and components of each shot (my Blender scene has 4 shots: a computer, phone, table, and board) 7. It took screenshots of every section, analyzed and fixed them 8. Synced every shot, zoom, and animation to the exact words I say 9. Animated a robot arm placing real logos on the table right when I mention them 10. Drew the whiteboard ending with logos and graphics instead of text 11. Pulled a cutaway from the creator's own video, with his voice 12. Rendered all 3,487 frames on RunPod cloud GPUs in under 20 minutes, for less than $1 13. Put me in the bottom third only when I'm talking, so the B-roll stays full screen 14. Added word-by-word captions, sound effects, and music 15. Uploaded it to my Google Drive, ready to post Will be making a full youtube video this week explaining it. I want to learn a little more about how cloud rendering works before making video, as well as figuring out what the best provider is. (If you are in cloud rendering space, hit me up) My goal of the youtube video will not be to make it a copyable skill, rather teach people the foundations so they can be creative and make their own mini-world by which they can make videos from, while understanding what's happening under the hood. Shout out Charlie Chapman Max Blade

Riley Brown

46,168 views • 23 days ago

Holy sh*t I spent weeks turning Claude Code into a design studio. no Figma license. no agency. no plugins I had to pay for. here's the stack I used and it's embarrassingly simple: -> graphify: teach Claude your brand dump your guidelines, logos, PDFs and old projects into one folder. graphify turns it into a knowledge graph, so every design Claude makes follows your rules instead of guessing. 134k stars, Y Combinator. uv tool install graphifyy -> graphify install -> /graphify . -> Figma-Context-MCP: Figma straight to code it hands your actual Figma file to the agent. the README says it builds designs in one shot, and does it better than pasting screenshots. -> Penpot: the open-source Figma free, open source, and its MCP server lets AI read and edit your designs, then hand them to code and back. -> React Bits: the "expensive" look, for free the biggest open-source library of animated React components. those smooth hover effects and text animations you see on high-end sites? copy, paste, done. -> HyperFrames: designs that move write HTML, get a finished video. made by HeyGen for AI agents. your landing page can now have its own launch video by tonight. claude plugin marketplace add heygen-com/hyperframes -> claude plugin install hyperframes@hyperframes -> MCP for Blender: 3D without learning Blender connects Claude to Blender. describe a scene, the agent builds and edits it in real 3D. the order that works: brand into graphify -> design in Figma or Penpot -> Claude builds it with React Bits -> HyperFrames makes the promo -> Blender adds the 3D hero shot the uncomfortable part: tools used to be the moat. now everyone has them for free. the only thing that still separates good design from AI slop is taste. and taste is the one thing you can't install. bookmark this. your next client will ask why you're suddenly so fast.

kiosa

73,242 views • 16 days ago