awesome-claude-video-skills
### Architecture & Positioning awesome-claude-video-skills is a curated repository and architectural design pattern designed to empower Claude Code, Cursor, and multimodal agents with end-to-end video understanding and processing capabilities. Handling video natively within text-centric LLM workflows faces two major roadblocks: massive token footprints that blow past context windows and the absence of structured pipeline orchestration for audio-visual media. This project bridges that gap by packaging modular skills that integrate FFmpeg media manipulation, Whisper audio transcription, and vision-language model keyframe analysis into standardized agent tools. ### Core Capabilities & Differentiation The primary strength of awesome-claude-video-skills lies in its dual-layer optimization: low-level media engineering and high-level multimodal synthesis. For long-form footage like conferences or technical lectures, the skills orchestrate intelligent scene-change detection and adaptive keyframe sampling, condensing hours of raw footage into dense, high-signal visual frames paired with time-aligned transcript text. For short-form and creative video pipelines, it provides structured workflows for shot-by-shot storyboard generation, highlight reel extraction, and automated subtitle translation. ### Deployment & Best Practices Integrating awesome-claude-video-skills requires a host environment equipped with FFmpeg built with hardware acceleration (e.g., Apple VideoToolbox or NVIDIA NVENC) for fast frame rasterization. To avoid redundant, costly video decoding on repetitive queries, pair these skills with a local multimodal vector store (such as LanceDB) to cache keyframe embeddings and timestamped transcript segments. For privacy-sensitive corporate meetings, route audio through local Whisper instances to keep raw multimedia data strictly within the corporate perimeter.
Features
Why choose it
Trade-offs
Maintenance check
- Last commit
- 2026-11-13
- Commits, last 90 days
- 57
- Releases, last 90 days
- 0
- Latest release
- —
- Open issues
- 19
- Recent issues with at least one comment
- 1/17
Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.
Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.
Most discussed open issues
- #1 Add: YeJe-cpu/SeeCut1 comments · 2026-09-28
- #21 Add: vericontext/vibeframe0 comments · 2026-11-13
- #20 Add: KyaniteLabs/kinocut0 comments · 2026-11-13
- #19 Add: kajisho5/ffmpeg-skill0 comments · 2026-11-13
- #18 Add: itpartypattaya/reels-montage0 comments · 2026-11-12
Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.
Compatibility
Use cases
Alternatives
Comments
No comments yet. Be the first!