{"id":51197078,"url":"https://github.com/louisedesadeleer/clipify","last_synced_at":"2026-06-28T04:01:03.273Z","repository":{"id":355848741,"uuid":"1229870384","full_name":"louisedesadeleer/clipify","owner":"louisedesadeleer","description":"Claude Code skill: turn long videos into social-ready clips. Auto-find funny moments, cut, reframe to 9:16 with face-tracking, and burn opus-style captions.","archived":false,"fork":false,"pushed_at":"2026-05-05T13:45:03.000Z","size":1135,"stargazers_count":12,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-05T15:38:29.323Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/louisedesadeleer.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-05-05T13:11:24.000Z","updated_at":"2026-05-05T15:38:16.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/louisedesadeleer/clipify","commit_stats":null,"previous_names":["louisedesadeleer/clipify"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/louisedesadeleer/clipify","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/louisedesadeleer%2Fclipify","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/louisedesadeleer%2Fclipify/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/louisedesadeleer%2Fclipify/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/louisedesadeleer%2Fclipify/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/louisedesadeleer","download_url":"https://codeload.github.com/louisedesadeleer/clipify/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/louisedesadeleer%2Fclipify/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34876271,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-28T02:00:05.809Z","response_time":54,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2026-06-27T21:31:06.299Z","updated_at":"2026-06-28T04:01:03.264Z","avatar_url":"https://github.com/louisedesadeleer.png","language":"Python","funding_links":[],"categories":["⭐ 精选第三方技能","Diagram and Visual Skills"],"sub_categories":["🎨 内容创作"],"readme":"# Clipify\n\nA [Claude Code](https://claude.com/claude-code) skill that turns long videos into social-ready clips.\n\n![Clipify turns long videos into 9:16 social clips](assets/preview.png)\n\nPoint it at any video file and it will:\n\n1. **Find clip-worthy segments** — transcribes the video with [Whisper](https://github.com/openai/whisper) and scans the transcript for punchlines, reversals, awkward pauses, and audio peaks to propose 3–5 candidates.\n2. **Create a 9:16 clip** — cuts your chosen moment, then reframes 16:9 → 9:16 with hard-cut pans that follow whoever is speaking (or split-screen if you'd rather see both faces).\n3. **Add subtitles** — burns opus-style word-by-word captions (big bold white, yellow active-word highlight).\n\nNo cloud APIs. Runs entirely on your machine. No OpenCV. ~20s of work for a 20s clip on Apple Silicon.\n\n## Why this exists\n\nMost \"auto-clip\" tools are either expensive SaaS, slow, or produce slop. This skill is what I actually use to clip my long-form videos for LinkedIn and TikTok. Built for talking-head dialogue (interviews, podcasts, two-person setups).\n\n## Requirements\n\n- macOS (uses VideoToolbox for hardware-accelerated decode — works on Linux/Windows if you remove `-hwaccel videotoolbox` flags)\n- [Claude Code](https://claude.com/claude-code)\n- `ffmpeg` with `libx264` (`brew install ffmpeg`)\n- [`whisper`](https://github.com/openai/whisper) (`pip install openai-whisper`)\n- Python 3 with `numpy` (`pip install numpy`)\n\n## Install\n\n```bash\ngit clone https://github.com/louisedesadeleer/clipify.git ~/.claude/skills/clipify\n```\n\nThat's it. Restart Claude Code and `/clipify` is available as a slash command.\n\n## Usage\n\nIn Claude Code:\n\n```\n/clipify\n```\n\nThen paste a video file path when asked. The skill will:\n\n1. Transcribe → propose 3–5 funny candidate clips with timestamps and titles\n2. Ask which to cut\n3. Ask 9:16 / 16:9 / 1:1\n4. If 9:16 from 16:9 with two faces: ask pan vs split-screen\n5. Ask subtitle style (opus / karaoke / minimal — or paste a reference image to match)\n6. Render and open the result\n\nFinal clips land in `\u003csource-video-dir\u003e/clipify_out/`.\n\n## How the face-pan works\n\nNo face detection model. Camera is static within a single clip, so:\n\n1. Eyeball each face's mouth+chin area as a rectangle on one sample frame.\n2. ffmpeg computes per-frame motion energy in each rectangle using frame differencing.\n3. Whichever rectangle has more motion at a given moment = that's the speaker.\n4. Build a hard-cut x-coordinate expression from the speaker timeline.\n5. Crop a vertical strip from the source that follows whoever's talking.\n\nTotal cost: a few seconds of ffmpeg per clip. Works surprisingly well.\n\n## Repo structure\n\n```\nclipify/\n├── SKILL.md           # the skill prompt Claude Code reads\n├── scripts/\n│   ├── analyze.py     # speaker timeline from two ROI motion files\n│   ├── build_pan.py   # ffmpeg crop x-expression with hard cuts\n│   ├── build_ass.py   # opus/karaoke/minimal ASS captions from whisper JSON\n│   └── audio_align.py # find offset of a sub-clip in a longer source\n└── README.md\n```\n\n## License\n\nMIT — see [LICENSE](LICENSE).\n\nBuilt by [Louise de Sadeleer](https://github.com/louisedesadeleer), Growth at [Tella](https://tella.tv).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flouisedesadeleer%2Fclipify","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Flouisedesadeleer%2Fclipify","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flouisedesadeleer%2Fclipify/lists"}