Remotion Agent Skills
remotion-dev/skills
Remotion's own skill set for coding agents that build videos as React code: project setup, markup rules, captions, live preview and rendering to video or stills.
- Editing
Pick one to let a coding agent cut footage from a transcript, render titles and captions from code, build a timeline in JianYing, CapCut or DaVinci Resolve, or handle dubbing, music and sound effects. Each entry says what it needs and links to its source.
75 skills
Start with where the edit will live, because that decides which skills matter to you. Code-built video skills have the agent write a project in React or HTML and render it, which suits titles, captions, explainers and launch films. Footage-cutting skills read a transcript or run FFmpeg to trim real recordings. Editor-driving skills build a draft or timeline in an app such as JianYing, CapCut or DaVinci Resolve, so you can finish by hand. The audio skills for dubbing, voice, music and sound effects can sit next to any of the three.
For code-built video, Remotion Agent Skills come from the Remotion team and cover project setup, markup rules, captions, live preview and rendering, with React as the medium. HyperFrames Skills come from HeyGen and do the same job in HTML, CSS and seekable animation under Apache-2.0, and they work in Claude Code, Codex, Cursor and Gemini CLI. Claude Remotion Skill is a smaller community add-on focused on motion design. Pick the one whose language your project already uses. Bigger projects such as OpenMontage and Claude Code Video Toolkit chain these tools with voice, music and image generation into a full pipeline. That takes more setup: OpenMontage needs Python, FFmpeg and Node.js, and some of the toolkit's AI tools need a Modal or RunPod GPU account.
For cutting real footage, video-use reads a word-level transcript instead of the video, proposes a cut plan, waits for your approval, then renders and checks its own cuts. The transcript comes from ElevenLabs Scribe, so you need an ElevenLabs API key. OpenEdit, from VEED, also cuts by transcript and builds captions and motion graphics as HTML; transcription can run on VEED's hosted service or locally with WhisperX, and billed generation calls ask you to confirm the price first. chengfeng-videocut is a Chinese-language Codex plugin for talking-head videos and needs a separate local Runtime. If you want nothing to leave your machine, ffmpeg-skill wraps 42 FFmpeg tools for cuts, captions, reframing, grading and exports, and needs only ffmpeg and Python.
If you want an editable timeline to finish by hand, choose a skill that drives the editor. JianYing Editor Skill scripts the desktop JianYing Pro app and builds a draft with subtitles, TTS voiceover, music and effects; its README says draft generation works on Windows and macOS, automatic export works only on Windows, and the international CapCut app is not supported. VectCutAPI Skills produce CapCut and JianYing drafts through the VectCutAPI service. DaVinci Resolve MCP skills cover grading, conform, Fairlight audio and delivery checks and are built for DaVinci Resolve Studio, with an in-app bridge described for the free edition. DaVinci AutoEdit Agent is a Codex pack that stops for your sign-off at each stage, and ButterCut exports timelines for Final Cut Pro, Premiere Pro and DaVinci Resolve.
Audio skills mostly wrap a vendor's API, so the skill itself is only text while the calls it makes are billed. ElevenLabs Skills cover speech, dubbing, transcription, music and sound effects and need an ElevenLabs API key. Sonilo Skills score a finished video, add matched effects and duck music under dialogue. ACE-Step Skills call a music model that you run yourself, and Violin video translator dubs a video into another language and writes subtitles.
Before installing any skill here, check five things. Which agents does the repo support? Some are Codex-only, others Claude Code-only. Which app or vendor does it target, such as JianYing Pro, Resolve Studio or an ElevenLabs account? What language is the skill text in? The JianYing skill is written in Chinese. What is the license? OpenMontage is AGPL-3.0, and ButterCut is noncommercial with an exception for the videos you make. And which paid accounts or GPUs does it need?
remotion-dev/skills
Remotion's own skill set for coding agents that build videos as React code: project setup, markup rules, captions, live preview and rendering to video or stills.
For developers and technical creators who want titles, captions and explainers as React code. These are the Remotion team's own skills, so the rules follow the framework; GitHub shows no license on the skills repo, so check the license terms before commercial use.
heygen-com/hyperframes
HeyGen's 21 skills for its open-source HTML-to-video framework: an agent plans, codes and renders explainers, launch videos, captions and music-synced clips as deterministic MP4s.
For people who would rather write HTML than React. HeyGen's 21 skills plan, build, preview and render explainers, launch videos, captions and music-synced clips as MP4 files in Claude Code, Codex, Cursor or Gemini CLI.
browser-use/video-use
A skill that lets a coding agent edit raw footage by conversation: it cuts from a word-level transcript, grades color, adds overlay animations and subtitles, then checks its own render.
For editing raw talking-head or tutorial footage by conversation. It cuts from a word-level transcript, asks you to approve the plan, then checks its own render; transcription needs an ElevenLabs API key.
kajisho5/ffmpeg-skill
A local FFmpeg editing skill with 42 scripted tools for cutting, captioning, reframing, grading, audio clean-up and platform exports, plus an MCP interface.
For anyone who wants edits to stay on their own machine. Its 42 scripted FFmpeg tools handle cuts, captions, reframing, grading, audio clean-up and platform exports, and it needs only ffmpeg and Python, with no account or API key.
luoluoluo22/jianying-editor-skill
Lets a coding agent script the desktop JianYing Pro app (CapCut's China edition): import media, add subtitles, TTS voiceover, music, effects and keyframes into an editable draft.
For people who finish their edits in JianYing Pro. It builds an editable draft with subtitles, voiceover, music and effects; automatic export works only on Windows, and CapCut users should look at VectCutAPI Skills instead.
samuelgursky/davinci-resolve-mcp
Agent skills shipped with a DaVinci Resolve MCP server, covering grading, editing, conform, Fairlight audio, delivery QC, Fusion, media-pool ingest, footage analysis, rough cuts and dead-air removal.
For colorists and editors who work in DaVinci Resolve. The skills route an agent through grading, conform, Fairlight audio, delivery checks, rough cuts and dead-air removal, and they ship with a Resolve MCP server.
elevenlabs/skills
Official ElevenLabs skills for the audio side of video: text-to-speech, dubbing, transcription, music, sound effects, voice changing and voice isolation.
For voice, dubbing, music and sound effects around a cut. They show an agent how to call ElevenLabs for speech, transcription, music, effects and voice isolation, so you need an ElevenLabs API key and the calls are billed by ElevenLabs.
calesthio/OpenMontage
An open-source repo in which a coding agent directs video production through staged pipelines, with a large set of skills on video-model APIs, music, speech, editing and composition.
heygen-com/hyperframes
HeyGen's 21 skills for its open-source HTML-to-video framework: an agent plans, codes and renders explainers, launch videos, captions and music-synced clips as deterministic MP4s.
browser-use/video-use
A skill that lets a coding agent edit raw footage by conversation: it cuts from a word-level transcript, grades color, adds overlay animations and subtitles, then checks its own render.
yikart/AiToEarn
Video-focused SKILL.md files from the AiToEarn content agent, led by a short-drama recap skill that submits clips for AI narration and voice-over; they rely on the platform's own tool calls.
remotion-dev/skills
Remotion's own skill set for coding agents that build videos as React code: project setup, markup rules, captions, live preview and rendering to video or stills.
luoluoluo22/jianying-editor-skill
Lets a coding agent script the desktop JianYing Pro app (CapCut's China edition): import media, add subtitles, TTS voiceover, music, effects and keyframes into an editable draft.
FireRedTeam/FireRed-OpenStoryline
A conversational video-editing agent whose editing workflows ship as skills, plus two skills that let Claude Code or OpenClaw install it and run an editing session.
samuelgursky/davinci-resolve-mcp
Agent skills shipped with a DaVinci Resolve MCP server, covering grading, editing, conform, Fairlight audio, delivery QC, Fusion, media-pool ingest, footage analysis, rough cuts and dead-air removal.
Agentchengfeng/chengfeng-videocut-skills
A Chinese-language Codex plugin with seven skills for editing talking-head videos: cuts, subtitles, overlay visuals, export and timeline edits, run through a local Runtime and Studio.
NarratorAI-Studio/narrator-ai-cli-skill
Lets an agent drive the Narrator AI command-line client to make narrated recap videos of films and short dramas, from choosing source material to a finished MP4.
sun-guannan/VectCutAPI
Two skills for agents that build editing drafts: one drives the VectCutAPI service for CapCut and Jianying, the other covers CapCut desktop draft files, vertical captions and audio mixing.
op7418/Youtube-clipper-skill
Claude Code skill that downloads a YouTube video, splits its transcript into chapters, cuts the segments you pick and can burn in bilingual Chinese and English subtitles.
MiniMax-AI/cli
Official skills for the MiniMax mmx command line: media generation, speech with subtitles, transcription, and a dedicated skill for running MiniMax-H3 video jobs.
digitalsamba/claude-code-video-toolkit
A Claude Code workspace with skills, slash commands and templates for explainer-style videos, covering Remotion, ffmpeg and moviepy composition plus AI voiceover, music, images and short clips.
Alisa0808/vox-director
Agent skill that makes a paper-collage explainer or ad video from a one-line topic, with collage keyframes, animation, narration, music and captions produced via Atlas Cloud models and ffmpeg.
0xsline/OpenChatCut
Four workflows bundled with the OpenChatCut video editor: AI short-film planning, shot-by-shot breakdown, beat-synced cutting and video model selection, loaded by its own agent or over MCP.
cclank/lanshu-create-ai-presenter-video
Takes a topic or script plus an authorized presenter photo through narration, presenter video, captions and a checked final render, with a state checker that gates each step on files on disk.
Agents365-ai/video-podcast-maker
Turns a topic into a narrated 4K explainer video: research, narration script, local text-to-speech, a Remotion composition with subtitles, and a render with background music mixed in by ffmpeg.
kajisho5/ffmpeg-skill
A local FFmpeg editing skill with 42 scripted tools for cutting, captioning, reframing, grading, audio clean-up and platform exports, plus an MCP interface.
veedstudio/open-edit
VEED's agent skill and CLI for editing video by prompt: it transcribes with word-level timing, cuts takes, builds captions and motion graphics as HTML, and renders an MP4.
feitangyuan/onetake
Builds 10 to 60 second product launch films and feature demos as one HTML composition, with measured motion moves, sound effects timed to on-screen events and an automatic continuity check.
higgsfield-ai/skills
Official Higgsfield skills that let an agent run image, video, 3D and audio generation through the Higgsfield CLI, build narrated explainer videos and train a reusable face identity.
shang-zhu/violin
A Claude Code skill that drives the Violin CLI to dub a video into another language and generate subtitles, using Together AI models and Cartesia speech synthesis by default.
feicaiclub/video-spec-builder
Interviews you about a video idea until it becomes a timed, shot-by-shot video-spec.md that the HyperFrames skill can render; it also edits an existing spec and checks changes for conflicts.
echris6/motion-video-kit
A Claude Code skill for planning, building and reviewing 15–40 second launch-style commercials that mix AI-generated footage with code-built motion, using an independent critic loop.
ChatCut-Inc/agent-plugin
Plugin from the ChatCut team that lets Claude Code, Codex and Cursor drive the hosted ChatCut editor, with skills for clip generation, voice and music, dubbing, captions, motion graphics and export.
eternityspring/reelbench-skills
Skills for studying finished videos: one breaks a video into a shot-by-shot table with measured durations, another renders it with the current shot's data beside the picture.
scenario-labs/skills
Scenario's skill pack for its MCP server, with video skills for Seedance, Kling, Veo and other models plus storyboards, consistent characters, assembly, captions and audio.
MapleShaw/seedance2.0-prompt-skill
A Chinese-language skill that writes video and image prompts for Seedance 2.0 on Jimeng, with a Z/Y/X/F camera code, a character-to-storyboard route for long videos and editing-rhythm formulas.
artokun/comfyui-mcp
Twelve skills from the comfyui-mcp plugin for video work in ComfyUI: a story-to-film pipeline, WAN 2.2, LTX-2.3 and MiniMax H3 workflows, clip extension, upscaling, color fixes and LoRA training.
op7418/guizang-product-video-skill
Builds software promo and release-notes videos from a product's real UI components, with GSAP or Three.js motion, a score and action sound effects, and a visual direction derived from the product.
Pluviobyte/video-production-skills
Four code-built video skills for HyperFrames or Remotion: a motion-first planning layer, a reference-video recreation checker, a dark SaaS promo style and a typed-text intro.
barefootford/buttercut
Agent skills for editing real footage: they analyze a library of clips, then build a scene, a selects reel or a rough cut and export a timeline for Final Cut Pro, Premiere Pro or DaVinci Resolve.
Jaycheng1103/chatgpt-video-editing-skills
Two skills: one checks and sets up a local editing environment, the other cuts your own footage into a vertical short in eight steps, with a transcript, preview and QA evidence along the way.
zenstory-ai/video-recap-skills
Six skills that turn one or several source videos into a Chinese-narration recap, covering scene analysis, scripting, voice-over, cutting, subtitles and ffmpeg assembly.
NoizAI/skills
Four audio and video-assembly skills from NoizAI: timeline-aligned text-to-speech, video dubbing, music-video assembly from a finished song, and text-to-sound-effect generation.
JimLiu/baocut
Lets an agent run the BaoCut app's command-line tool to transcribe audio or video, build and translate subtitles, edit timelines and overlays, and export the result.
liuluhaixiu/DaVinci-AutoEdit-Agent
A Codex skill pack that takes a folder of footage through a script, an edit blueprint and an audited DaVinci Resolve timeline, pausing for your sign-off at each stage.
elevenlabs/skills
Official ElevenLabs skills for the audio side of video: text-to-speech, dubbing, transcription, music, sound effects, voice changing and voice isolation.
heygen-com/skills
Official HeyGen skills that let an agent create a reusable avatar, generate presenter-led videos through HeyGen's Video Agent, and dub finished videos into other languages.
geekjourneyx/hyperframes-motion-director
Directs HyperFrames videos in two phases, first a brief for your approval, then assets, composition, render and review, with Chinese 9:16 promos as the default.
joeseesun/qiaomu-cut-skill
A video-director skill: from a one-line brief it plans the project, sources footage, builds subtitles and narration, draws motion graphics in code and renders with ffmpeg, with checks.
Mr-funny/hbg-classical-poem-silk-video
Turns a Chinese classical poem into a vertical 9:16 Chinese-painting video: scenes per verse, generated stills, image-to-video clips, brush-calligraphy captions and a final QA pass.
hi-nikola/hand-drawn-explainer-video-nikola
A Chinese-language Codex skill that turns a topic, script or SRT into a hand-drawn explainer video with narration, subtitles, a timeline and a verified MP4.
aaronyi97/image-story-video-wizard
A confirmation-gated skill that guides you from topic through script, voiceover, storyboard and generated images to a rendered narrated image-story video, such as an AI book talk or audio story.
op7418/Video-Wrapper-Skills
Reads an interview video's SRT subtitles, proposes on-screen graphics such as lower thirds and keyword callouts for approval, then renders them onto the video in one of four themes.
Bomx/super-video-maker-skill
An end-to-end skill that runs a video idea through script, asset generation, assembly, FFmpeg finishing and QC, using HeyGen avatars, Seedance b-roll, OpenAI images, Remotion or HyperFrames.
hassancs91/claude-faceless-shorts-creator
Five Claude Code skills that build faceless vertical YouTube Shorts: Remotion-coded shorts, character-locked clips from a fal video model, Vox-style collage shorts and sound design.
haidrrrry/claude-remotion-skill
A community Claude skill for building Remotion videos with deliberate motion design: eased and spring animation, staggered entrances, layered backgrounds and a render-and-inspect loop.
dean9703111/ai-agent-skill-for-video-workflow
Four skills for the subtitle side of video work: transcribe audio to SRT, correct it against a script, add annotation cards under the lines and draft platform-specific social posts.
Comfy-Org/comfy-skills
Comfy's own plugin and skill for driving the hosted Comfy Cloud MCP server from an agent: image, video, audio and 3D generation, ComfyUI workflow templates, and model, node and template search.
volcengine/mediakit-cli
Official Volcengine skills for mediakit-cli: cut and mix video and audio, extract subtitles, and turn finished short-drama episodes into scripts, storylines, highlight reels and recap videos.
YeJe-cpu/SeeCut
Auto-edits talking-head or AI-avatar footage into a motion-heavy short video, with Gemini watching each render and pairwise comparison choosing between versions.
calesthio/generative-media-skills
A large skill suite for AI media production: video-model operating guides, shot and storyboard direction, character continuity, editing and delivery checks across image, video and audio.
Cassette-Editor/oh-my-cassette
An editing plugin and local MCP server that lets an agent trim, cut, caption, add music and export video through the hosted Cassette editor.
Mr-funny/hbg-life-simulation
Turns a long Chinese life story into a narrated video of still comic frames: locked protagonist, flash-cut opening, TTS-aligned captions, slow zoom and pan, rendered in HyperFrames.
znyupup/ai-video-editing-skill
Workflow for an agent to turn a folder of raw travel clips into a finished vlog using ffmpeg, speech recognition and a vision model, with a browser preview before rendering.
jianshuo/claude-skills
Thirteen Claude Code skills for video post-production: transcription, subtitle translation, dubbing, multicam sync and cutting, clip segmenting, reframing, HyperFrames overlays and YouTube upload.
black-forest-labs/skills
Official Black Forest Labs suite of eight linked skills for FLUX 3 video: router, brief checker, prompt craft for shots, keyframes, audio and period looks, API runner and product-ad workflow.
sonilo-ai/skills
Agent skills for the Sonilo audio API that score a finished video with music, add matched sound effects, dub speech into other languages and duck music under dialogue.
linyqh/speclip-skills
Skills for short-drama commentary videos, talking-head rough cuts, FFmpeg export commands and subtitle layout on portrait and landscape video.
ace-step/ace-step-skills
Skills that let an agent write songs, generate music through a running ACE-Step 1.5 API, transcribe lyrics to LRC and render a simple lyric music video with Remotion.
MCKRUZ/ComfyUI-Expert
Claude Code project that routes video requests to ComfyUI skills for prompts, character identity, video engines, voice and lip-sync, and ffmpeg or Remotion assembly on a ComfyUI install.
shengshu-ai/vidu-skills
Official Vidu skill that lets an agent submit text-to-video, image-to-video, start/end-frame and reference-based jobs, plus text-to-speech, lip-sync and compose, through the vidu-cli command line.
runwayml/skills
Official Runway skills that run scripts against the Runway API to generate text-to-video, image-to-video and video-to-video clips, plus speech, sound effects, voice isolation and dubbing.
PixVerseAI/skills
PixVerse's own skill library that teaches an agent to drive the PixVerse CLI for video, image, speech and music generation, with Seedance prompt optimizers and reusable character references.
geegl/directorskills
A library of 98 skills: 97 distilled from ten film-craft books and one cross-book camera reference, covering story, directing, shot design, cinematography, color grading, transitions and sound.
guigulaoshi/music-video-director-skill
Cuts a music video from a song, source footage and a one-line director's note: analyzes beats, lyrics and scenes, drafts an edit list for approval, then renders it with ffmpeg via a bundled CLI.
eachlabs/skills
Eight video-focused skills from each::labs' own repository that call its hosted API for video generation, subtitles, color grading, lip sync and localization.
PostPlusAI/postplus-skills
Skills for the PostPlus CLI covering batch video generation, shot-by-shot reference analysis, reference decoding, transcription to SRT or ASS, and voice planning and execution.
cyuanxv/ai-mandrama-skills
Two Chinese-language Claude Code skills for AI comic-drama and short-drama episodes: a nine-stage pipeline built on the Dreamina CLI and ffmpeg, and Chinese multi-character voice-over with edge-tts.
vargHQ/skills
Skill for the varg platform that generates video, images, speech and music from prompts and composes them into captioned clips, through a hosted render API or a local bun and ffmpeg setup.
opus-pro/opus-skills
Official OpusClip skill that cuts long videos into short clips via the OpusClip API, with brand templates, captions, dubbing and scheduled posting to connected social accounts.
runcomfy-com/skills
RunComfy's own skills for its hosted model API: a video-model router plus single-model, lip-sync, video-edit, extend, inpaint and outpaint skills run through the runcomfy CLI.
magichourhq/skills
Magic Hour's own skill pack for AI video work: image-to-video, text-to-video, character consistency and replacement, music and talking videos, editing, remixes, explainers and review.
Run npx skills add remotion-dev/skills in your project. The README says the same pack is also offered when you create a project with bun create video. Then describe the video you want: the router skill, remotion-best-practices, hands the request to the sub-skill for project setup, markup, captions, preview or rendering, and it opens a Remotion Studio preview rather than rendering by default. The pack is meant for Codex and Cursor too. Details are on the Remotion Agent Skills page.
Yes, in three ways, and a skill tells the agent which one to use. It can write code that renders a video with Remotion or HyperFrames, run FFmpeg to cut and caption real footage, or script an editor such as JianYing or DaVinci Resolve to build a timeline. Skills like video-use and ffmpeg-skill add the procedure, which is to inspect the files, edit, then verify the result. Expect a cut that you review and adjust.
Both let an agent write a video as code and render it to MP4. Remotion works in React and its skills come from the Remotion team; HyperFrames works in HTML, CSS and seekable animation, and its skills come from HeyGen. HyperFrames also ships a skill for porting an existing Remotion composition to HTML. Choose by the language your project already uses, then compare Remotion Agent Skills with HyperFrames Skills.
Yes. ffmpeg-skill installs with npx ffmpeg-skill, and its README covers Claude Code, Cursor and Codex. It ships 42 scripted tools for cutting, captions, reframing, color grading, audio clean-up and platform exports, plus an MCP interface, and it runs offline with ffmpeg and Python 3 on your PATH. Volcengine MediaKit Skills also run many editing operations on local FFmpeg, though their cloud tools need a Volcengine API key.
Most skills are text files in public repositories with no price of their own, but many call paid services. video-use needs an ElevenLabs API key for transcription, ElevenLabs Skills and Sonilo Skills call billed APIs, and Claude Code Video Toolkit needs a Modal or RunPod account for some of its AI tools. For a fully local setup, ffmpeg-skill needs no key, and ACE-Step Skills use a music model you run yourself. Licenses differ too, so read each repo before you build a commercial workflow on it.
REEL END — 00:00:00:00
Pencil0 runs the whole path for you: script, assets, shots, then a final cut you can screen. Invite-only beta.
Join waitlistFree to join · we'll email you the day we open