MCP App StoreMCP App Store
from claude-shorts70
Analyze long videos, extract high-impact short clips, clean captions, and render platform-ready vertical videos with animated captions and optimized exports.
Shorts is an interactive pipeline skill that turns long-form videos into platform-optimized short clips. It transcribes audio (faster-whisper), scores candidate segments with an AI rubric, presents selections for user approval, snaps boundaries to natural audio cut points, and renders animated captions via Remotion before exporting platform-specific encodings.
Use when you want to extract viral-ready clips from longer videos (talking-heads, podcasts, screen recordings) while keeping human approval in the loop. Suitable for content creators who need captioned, reframed, and encoded vertical videos for YouTube Shorts, TikTok, and Instagram Reels.
Designed for Claude Code-style agents or any agent that can run Bash and Python scripts and interactively ask user questions. It relies on local tooling (ffmpeg, node/Remotion, Python).
A comprehensive 10-step interactive pipeline for converting longform video into shortform content (YouTube Shorts, TikTok, Instagram Reels). Scripts are well-structured Python/Bash with proper CLI argument parsing and JSON output. Only detect_gpu.sh ran successfully (no GPU on this machine); other scripts require input video files and correctly produced usage errors. Heavy dependency chain (faster-whisper, mediapipe, opencv, Remotion/Node) makes initial setup non-trivial.
faster_whispermediapipeopencv-pythonremotion (node)No security concerns. Scripts operate entirely locally on user-provided video files. No network calls, no credential handling, no data exfiltration. The SKILL.md is exceptionally detailed — one of the best-structured skills seen. The 10-step pipeline with interactive approval (Step 5) and boundary snapping (Step 7) shows thoughtful UX design. Shell scripts use set -euo pipefail consistently.
NotebookLM Research Agent
End-to-end research-to-content pipeline: ingest sources into Google NotebookLM, run deep cited research, generate downloadable artifacts, and produce polished c
Learn: From Raw Materials to Published Output
Run a disciplined six-phase research-to-publish workflow (collect, digest, outline, fill, refine, publish) for deep domain research and article preparation.
Filtrix Video Gen (MCP)
Submit and manage text- or image-to-video generation jobs via Filtrix's Remote MCP API, including polling and downloading completed videos.
Voice Memo Organizer
Locate, transcribe (local whisper.cpp), summarize and index Apple Voice Memos into a searchable archive with titles, themes and key quotes.
FFmpeg Guide
Comprehensive FFmpeg reference for encoding, converting, streaming, filtering, and analyzing audio/video — command examples, common patterns, and troubleshootin
TensorsLab Image Generation
Generate and edit images (text→image, image→image, advanced edits) using TensorsLab models with automatic prompt enhancement and local output saving.
Provide Raw YouTube Video ID
Return only a valid existing YouTube video ID string (no URL, no commentary) — useful for tools that need raw IDs for downstream operations.
YouTube Tools (yt-dlp)
Free, local YouTube operations using yt-dlp: download videos/audio, extract transcripts/subtitles, and fetch metadata (no API keys).