FFmpeg plugins

Create videos programmatically by writing HTML, CSS, and JavaScript with HyperFrames. Generate talking-head captions, faceless explainers, motion graphics, slideshows, product launches, and website captures. Includes CLI for scaffolding, preview, and cloud rendering on AWS Lambda.

Record, transcribe, and search meetings and voice memos locally. Query conversation history, extract decisions and action items, detect conflicts across meetings, and generate daily or weekly briefs. Includes debrief analysis, meeting prep with relationship context, self-coaching metrics, and product walkthrough review.

Voice conversations with Claude Code using local speech-to-text and text-to-speech, enabling hands-free interaction, custom voice cloning, remote voice sessions via cloud platform, and background music control during voice sessions.

Downloads videos from URLs or local paths, extracts frames at an auto-scaled rate, pulls captions or falls back to Whisper transcription, and lets Claude answer questions grounded in both the visual frames and transcript.

Enable Claude to analyze local video files or YouTube URLs with full multimodal perception: extract frames and audio via FFmpeg, detect scenes/motion/silence/transitions, transcribe speech using Whisper or Gemini, generate detailed visual descriptions, summaries, and answer questions interactively.

Automate legal document workflows: convert PDFs/scans to Markdown, transcribe audio/video, generate contracts and patent drafts, manage case files, and coordinate multi-agent research—all with Chinese legal domain optimization.

Turn any video into a Chinese-narrated recap using ffmpeg and one MiMo API key: understand content via ASR and VLM, write a timestamped narration script, cut source clips, synthesize Chinese TTS voiceover, and assemble the final video with subtitle rendering, audio ducking, and loudness normalization.

Process videos using natural language commands: ingest from files, URLs, RTSP feeds, or desktop capture; index visual and spoken content for search; transcode, edit timelines, generate assets, and create real-time alerts via VideoDB Python SDK.

Provides 18 universal AI skills for Claude Code, enabling skill creation and optimization, deep multi-source research with citations, strategic planning using business frameworks, content processing (audio/video transcription, document conversion, web scraping), software architecture design with C4 diagrams, and orchestrated multi-step execution with review checkpoints.

Downloads videos from URLs or local paths, extracts scene-change frames and pacing metrics, transcribes via captions or Whisper, and produces a structured editorial analysis report with optional Obsidian vault integration.

Perform deep research on videos and web content using Gemini, then automatically generate structured reports, knowledge graphs, and explainer videos with narration and AI-generated visuals.

Extract key frames, audio transcripts, and metadata from videos and animated images into a self-contained HTML report that Claude can view, hear, and annotate with analysis.

Generate professional videos from text scripts using AI voiceovers, stock footage, and Remotion rendering, with CLI commands for running, resuming, and debugging video generation workflows.

Generate technical presentations, courses, marketing content, changelogs, architecture impact docs, and promo videos from code changes or freeform input, orchestrated as a sequenced workflow for AI coding agents.

Automate video editing and rendering with FFmpeg and Remotion by stitching clips, adding transitions, captions, teasers, and Whisper transcription. Generate professional AI infographics from natural language or content analysis using Gemini. Conversationally design Excalidraw presentations, diagrams, and slide decks, exporting JSON or injecting into excalidraw.com via browser automation.

Orchestrate multi-platform marketing campaigns from Claude using the Hyper MCP — manage paid ads on Google, Meta, TikTok, Pinterest, Amazon, LinkedIn, plus organic social, SEO research, email outreach, competitor analysis, ad creative generation, and analytics integrations.

Write, analyze, and optimize blog content with anti-hallucination verification, SEO validation, multilingual translation, AI citation readiness, and quality scoring across 34 sub-skills and 8 agents.

Automates multi-platform content publishing and asset generation, including WeChat stickers, Toutiao articles, tutorial videos, and Excalidraw diagrams via CLI or browser automation.

Automate video processing workflows: download, trim, remove silence, extract audio, transcribe with Whisper, generate captions, change speed, concatenate, enhance audio, and upload to YouTube or Bunny.net

Download videos from 1800+ platforms like YouTube, TikTok, and Twitter, then auto-generate MP4 files, MP3 audio extracts, VTT subtitles, TXT transcripts, and Markdown AI summaries. Everything saves directly to your downloads folder for easy access and reuse in projects.

Generate deterministic motion-graphics videos as editable TypeScript scenes; scenes are pure functions of time, enabling data-driven batch renders of title cards, lower thirds, kinetic typography, and product teasers that survive AI regeneration.

Generate accurate FFmpeg commands from natural-language descriptions, debug failed runs, and optimize encoder settings with ffprobe-based analysis — no need to memorize complex flags or filter syntax.

Generate visuals across 11 channels including raster, SVG, web, interactive, video, terminal, diagrams, and infographics while enforcing brand tokens, 8:1 contrast minimum, custom dimensions, and style guides. Critique produced PNG/JPG/SVG/PDF artifacts with a vision-model agent that evaluates design principles (Tufte/Bertin/Gestalt), channel anti-patterns, and brand compliance, providing evidence-based PASS/NEEDS REVISION/FAIL verdicts.

Enforces a TDD-first team workflow: every change starts as a plan with tests, then iteratively passes those tests before archiving to a regression suite. Reports progress to Slack/Discord and skips obsolete tests via supersede chains.

Record TikTok-style MP4 videos walkthroughing your local dev server UI on ports like 3000/5173/8080, scaffold tiktest.md configs automatically from README/package.json, run checks-only validation passes without rendering, and summarize changes via tik-test CLI.

Generate AI voice content and videos from text via CLI: synthesize speech in 200+ voices across 40+ languages, create multi-speaker podcasts, transcribe audio/video with word-level timestamps, dub videos from SRT, translate videos end-to-end, and turn articles into vertical card videos or shareable card images.

Turn raw content ideas into a full marketing engine: draft LinkedIn/X posts, YouTube scripts, thumbnails, newsletters, lead magnets, and motion videos, all in a consistent brand voice, tracked in an Obsidian vault.

Develop, debug, and deploy Home Assistant automations, custom Python integrations, Lovelace dashboards, Docker add-ons, Assist voice control, energy monitoring setups, and REST/WebSocket API connections using specialized skills, commands, and agents.

Extract key frames from videos using ffmpeg scene detection, transcribe audio with optional Whisper, and analyze screen recordings, bug reports, tutorials, and demos directly in Claude.

Accelerate game development in Godot/Summer Engine with AI-driven asset generation (2D/3D sprites, models, animations, audio, VFX), scene composition, multiplayer networking, performance profiling, debugging, and release validation — all through natural language commands and an MCP bridge to the local desktop app.

Download videos from YouTube, Instagram, X, Vimeo, TikTok, or local files, then extract frames and generate local transcripts (via mlx-whisper) so Claude can answer questions about the video content without sending data off-device.

Turn Claude Code into a professional media production workstation: transcode, stream, package, QC, color-grade, and deliver video/audio across broadcast, OTT, and AI-enhanced pipelines using FFmpeg, OBS, GStreamer, WebRTC, and 90+ open-source media tools.

Orchestrate multi-agent development workflows with constructor-pattern primitives: scaffold APIs, docs, CI/CD, and auth; run parallel code audits, debugging, and quality gates; generate animations, images, and social content; manage cross-session messaging and cloud consolidation.

Drive autonomous iteration loops for research tasks, generate publication-quality data visualizations and scientific code, fetch arXiv full-text and BibTeX citations, create images via Gemini, and stream text-to-speech.

Convert QQ Music .ogg/.oggh files to MP3 with embedded cover art and metadata on macOS, batch-process directories, inspect file metadata, and check environment dependencies—all via slash commands or natural language.

Select, implement, review, and migrate Rust FFmpeg libraries like ez-ffmpeg and ffmpeg-next to build video/audio processing pipelines, handle media conversion, and tackle multimedia tasks with guided code generation and problem-solving workflows.