Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
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.
npx claudepluginhub damionrashford/media-os --plugin media-osBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Designs end-to-end media pipelines before a single command runs. Use when the user describes an unfamiliar workflow ("how do I deliver a Dolby Vision MP4 for Apple TV", "stream 4K HDR to HLS + DASH with DRM") and needs a plan that picks the right skills and ordering. Produces a staged recipe with exact tools, not just prose.
Packages finished renders for streaming distribution and VOD — HLS/DASH packaging with Shaka/GPAC, MP4 fragmentation, DRM (Widevine/PlayReady/FairPlay) with cbcs, CDN upload (Cloudflare Stream, Mux, Bunny), and platform-spec delivery (YouTube, Vimeo, broadcast IMF/MXF). Use when the encode is done and the file needs to get to viewers.
Executes video/audio encodes with the right codec, container, and flags for the target. Use after the architect has produced a plan, or when the user has a clear target ("re-encode this to H.265 10-bit under 5 Mbps", "transcode to ProRes 422 HQ for Premiere"). Owns rate-control math, preset selection, pixel format, and container flags.
Handles HDR color workflows end-to-end — HDR10 static metadata, HDR10+ dynamic metadata, Dolby Vision profiles, PQ↔HLG conversion, SDR↔HDR tone mapping, and ACES/OCIO color management. Use when the user mentions Dolby Vision, HDR10, HDR10+, HLG, PQ, tone mapping, or anything about color pipelines that must survive delivery.
Runs live production — OBS control over obs-websocket, multi-bitrate streaming (RTMP/SRT/RIST/WHIP), NDI routing, DeckLink capture, PTZ camera control, and low-latency contribution feeds. Use when the user is LIVE — streaming now, about to go live, mid-show, or wiring up the broadcast rig.
Manipulate macOS audio at the Core Audio HAL layer: afinfo (audio file metadata), afplay (play), afconvert (convert format/rate/bit-depth/channel layout), say (TTS), SwitchAudioSource (change default input/output/system device — brew install switchaudio-osx), audiodevice CLIs (npm/Go variants), AppleScript via osascript to drive System Preferences Sound. HAL plugins at /Library/Audio/Plug-Ins/HAL/ load virtual drivers (BlackHole 2ch/16ch/64ch, Loopback, Background Music). Aggregate Device (drift-corrected multi-in) and Multi-Output Device via Audio MIDI Setup.app or AudioHardwareCreateAggregateDevice. Use when the user asks to change default macOS audio device, route audio between apps on Mac, install BlackHole, build an Aggregate Device, convert audio files with afconvert, or script Mac audio from the terminal.
Route professional low-latency audio with JACK Audio Connection Kit (jackaudio.org): jackd server (ALSA/CoreAudio/PortAudio/ASIO backends), jack_control (D-Bus JACK2 config), jack_lsp (list ports/connections), jack_connect/jack_disconnect (manage links), jack_cpu_load, jack_samplerate, jack_bufsize (runtime period size), jack_transport, jack_rec (multi-channel WAV recording), jack_iodelay (round-trip latency), jack_wait, jack_midi_dump. JACK1 C maintenance vs JACK2 C++ multicore. Modern Linux swaps jackd for PipeWire's libjack shim — same client API. JackTrip for uncompressed multichannel UDP over the Internet. Use when the user asks to run a JACK server, connect audio clients, measure roundtrip latency, record from JACK ports, set JACK period size, or bridge JACK over the network with JackTrip.
Route and introspect Linux audio/video/MIDI graphs with PipeWire: pw-cli (interactive object/module control), pw-dump (full graph JSON with --monitor), pw-link (list/create/destroy port links), pw-cat / pw-play / pw-record (playback+record), pw-top (real-time DSP load + xrun viewer), pw-metadata (shared metadata store), pw-loopback (source/sink bridges), pw-midiplay/pw-midirecord, pw-jack (JACK client compat shim), pw-mon, pw-profiler. WirePlumber session manager handles routing/defaults/profiles. Replaces PulseAudio (via pipewire-pulse) and JACK (via libjack.so shim) simultaneously. Use when the user asks to route Linux audio, link two apps, list audio devices on Linux, create a loopback, record system audio, monitor the audio graph, or set up a virtual audio device on modern Linux.
Manipulate Windows audio via WASAPI (learn.microsoft.com/coreaudio): SoundVolumeView + svcl.exe (NirSoft — set default device, per-app volumes, save/load profiles), AudioDeviceCmdlets PowerShell module (Get-AudioDevice, Set-AudioDevice, Get-AudioDevicePlaybackMute), nircmd setdefaultsounddevice. Shared mode (Windows engine resamples all streams to endpoint fixed format) vs Exclusive mode (bit-perfect, audiophile, ~sub-3ms latency with IAudioClient3 + event-driven callbacks). WASAPI then MMDevice then Audio Engine (audiodg.exe) then WDM/KS. ASIO is Steinberg's parallel stack bypassing Windows audio. Virtual audio devices: VB-Audio Virtual Cable, VoiceMeeter/Banana/Potato. Use when the user asks to change Windows default audio device, route app audio, use VoiceMeeter, install VB-Cable, set a specific device from PowerShell, or enable exclusive-mode low-latency output.
Google MediaPipe Tasks API (current supported, NOT legacy mediapipe.solutions): Vision Tasks (Face Detector, Face Landmarker, Face Stylizer, Hand Landmarker, Gesture Recognizer, Pose Landmarker, Holistic Landmarker, Object Detector, Image Classifier, Image Segmenter, Interactive Segmenter, Image Embedder, Image Generator), Audio Tasks (Audio Classifier), Text Tasks (Text Classifier, Text Embedder, Language Detector), GenAI Tasks (LLM Inference — on-device Gemma/Phi-2/Falcon/StableLM). Three run modes: IMAGE, VIDEO, LIVE_STREAM (async). Python: mediapipe.tasks.python.vision/audio/text/genai. Docs at ai.google.dev/edge/mediapipe/solutions. Model Maker for transfer learning. Use when the user asks to detect pose/face/hand landmarks, classify an image, run a body-pose model, segment people, recognize gestures, do on-device LLM inference via MediaPipe, or build a mobile/embedded CV pipeline.
Probe SHAKA_KEY_SERVER_URL; warn on auth failures or extended downtime.
Poll nvidia-smi for temperature + sustained-load warnings so Claude knows before the GPU throttles.
Poll INCOMING_MEDIA_DIR for new stable media files; notify Claude with a probe suggestion.
Poll RENDER_QUEUE_URL (HTTP JSON) or RENDER_QUEUE_DIR (status sidecars); notify on job state transitions.
ffprobe LIVE_STREAM_URL every STREAM_POLL_SEC; notify on down/up/bitrate-drop transitions.
This plugin requires configuration values that are prompted when the plugin is enabled. Sensitive values are stored in your system keychain.
MUX_TOKEN_IDMux API token ID for media-cloud-upload direct uploads.
${user_config.MUX_TOKEN_ID}MEDIA_WORK_DIRScratch directory for intermediate renders, proxies, and QC reports. Must be writable and have enough space for large media.
${user_config.MEDIA_WORK_DIR}BUNNY_CDN_TOKENBunny.net Stream API token for media-cloud-upload direct uploads.
${user_config.BUNNY_CDN_TOKEN}LIVE_STREAM_URLLive stream contribution URL (RTMP/SRT/RTSP/HLS) for the stream-health monitor to probe. Empty disables.
${user_config.LIVE_STREAM_URL}MUX_TOKEN_SECRETMux API token secret for media-cloud-upload direct uploads.
${user_config.MUX_TOKEN_SECRET}FFmpeg CLI reference for video and audio processing, format conversion, encoding, filtering, and media automation
Helps Claude Code generate accurate, efficient FFmpeg commands from natural-language descriptions.
Claude Code skill pack for Runway (18 skills)
Complete FFmpeg expertise system with comprehensive coverage of FFmpeg 8.0.1 (2025-11-20) features, VIRAL VIDEO CREATION for TikTok/YouTube Shorts/Instagram Reels/Twitter, KINETIC CAPTIONS (word-grow karaoke, bounce, pop, elastic, spring physics), hardware acceleration (NVIDIA NVENC, Intel QSV, AMD AMF, VAAPI, Vulkan), Docker/container usage, Modal.com serverless containers (CPU/GPU), Cloudflare Workers/WebAssembly (ffmpeg.wasm), CI/CD runner optimization (GitHub Actions), live streaming (RTMP/HLS/DASH/SRT/WebRTC/WHIP), audio processing and normalization, encoding optimization (CRF, presets, quality tuning), FFmpeg + OpenCV integration, PyAV Python bindings (FFmpeg 8.0+ compatible), and common gotchas/troubleshooting. PROACTIVELY activate for: (1) ANY FFmpeg task (encoding, transcoding, streaming, filtering), (2) VIRAL VIDEO creation (TikTok, YouTube Shorts, Instagram Reels, Twitter/X), (3) Short-form vertical video (9:16 aspect ratio), (4) Auto-captions and animated subtitles (CapCut-style), (5) KINETIC CAPTIONS with word-grow, bounce, pop, elastic effects, (6) Karaoke-style word highlighting with animations, (7) Video hook optimization (first 3 seconds), (8) Multi-platform batch export, (9) Social media video optimization, (10) Video format conversion (MP4, WebM, MKV, AVI), (11) Hardware acceleration setup (NVENC, CUDA, QSV, VAAPI, AMF, Vulkan), (12) Docker container FFmpeg usage (jrottenberg/ffmpeg, linuxserver/ffmpeg), (13) Modal.com serverless video processing (GPU/CPU containers, parallel processing, volumes), (14) Browser/WebAssembly FFmpeg (ffmpeg.wasm), (15) Cloudflare Workers video processing, (16) GitHub Actions/CI pipeline video processing, (17) Live streaming setup (RTMP ingest, HLS/DASH/WHIP/WebRTC delivery), (18) Audio encoding/normalization (AAC, MP3, loudnorm, EBU R128), (19) Quality optimization (CRF, bitrate, presets), (20) FFmpeg debugging and error resolution, (21) VVC/H.266, AV1, HEVC encoding, (22) Filter graphs and complex filtergraphs, (23) Batch processing and automation, (24) Whisper AI subtitle generation, (25) Content creator workflows, (26) FFmpeg to OpenCV frame handoff, (27) cv2.VideoCapture vs ffmpeg subprocess, (28) BGR RGB color format conversion gotchas, (29) Frame dimension order img[y,x] vs img[x,y], (30) ffmpegcv GPU-accelerated video I/O, (31) VidGear multi-threaded streaming, (32) Decord batch video loading for ML, (33) PyAV frame-level processing (av.open, VideoFrame, AudioFrame, filter graphs), (34) Pipe frames between FFmpeg and OpenCV, (35) OpenCV + FFmpeg + Modal parallel processing, (36) PyAV installation on Ubuntu 20.04/22.04/24.04, (37) Building PyAV against custom FFmpeg, (38) PyAV FFmpeg 8.0+ compatibility and features. Provides: KINETIC CAPTIONS skill (word-grow karaoke, spring physics, elastic bounce, platform-specific timing), VIRAL VIDEO skills (platform specs, hook templates, animated captions, batch social export), PyAV integration skill (installation guides, FFmpeg 8.0+ compatibility, av.open patterns, VideoFrame/AudioFrame NumPy conversion, filter graphs, encoding/decoding, seeking, RTSP streaming, memory management, thread safety, error handling), FFmpeg 8.0.1 patch release features, FFmpeg 8.0 Huffman features (Whisper AI filter, Vulkan compute codecs for FFv1/ProRes RAW/AV1/VP9, APV codec, VVC VA-API/QSV decoding, WHIP muxer for WebRTC, colordetect/pad_cuda/scale_d3d11 filters, ProRes RAW decoder, RealVideo 6.0, G.728), FFmpeg 7.1 LTS features (VVC decoder, MV-HEVC for Apple Vision Pro, xHE-AAC, Vulkan H.264/H.265), hardware acceleration guides, Docker best practices, Modal.com parallel processing patterns, WebAssembly limitations and workarounds, CI/CD optimization strategies, streaming protocol guidance, audio processing workflows, encoding presets and recommendations, systematic debugging approaches, FFmpeg + OpenCV integration patterns, BGR/RGB color conversion, ffmpegcv/VidGear/Decord/PyAV library guidance, and production-ready 2025-2026 configurations for all platforms.
Agent skill and CLI helpers for using RivalSearchMCP research tools from Claude Code. Includes workflows, command references, and a standalone CLI for the hosted MCP server.
9 research tools: 5-engine web search, 9 social platforms, 5 news sources, 5 academic databases, GitHub, website mapping, document analysis, and research topic synthesis. No API keys required. Deterministic outputs for agent chaining.
Critical thinking platform for decision analysis, proposal review, debate preparation, and due diligence. Three agents (advocate, adversary, judge) with a structured argument graph, sequential thought chain, and 10 reasoning traditions.
ML Workbench for Claude Code. Full ML lifecycle: search papers across 7 academic sources, discover and download datasets from 5 repositories, explore and clean data, engineer features, train models (Naive Bayes, KNN, LDA/QDA, SVM, Decision Trees, Ensembles, GLM, Gaussian Process, Neural Networks), run autonomous experiments, build AI apps with LLMs and RAG, build MCP servers, deploy models with Docker and CI/CD, detect drift, explain predictions with SHAP, generate podcasts from papers, manage notebooks, extract YouTube content, and learn ML interactively with 3 university-grade courses (CS229, Applied ML, ML Engineering). 11 agents, 16 skills, 3 CLI tools (mlx-exp, mlx-search, mlx-status), 1 MCP server, 3 output styles, Python LSP via pyright.
Elite research intelligence plugin powered by RivalSearchMCP. 10 research tools, 5 workflow skills, 6 specialist agents, portable hooks, zero API keys. Your AI research team.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimExecutes bash commands
Hook triggers when Bash tool is used
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
RENDER_QUEUE_DIRRender-queue status-sidecar directory (*.status files) as an alternative to RENDER_QUEUE_URL. Empty disables.
${user_config.RENDER_QUEUE_DIR}RENDER_QUEUE_URLRender-farm status endpoint (JSON list of {id,status,name}) polled by the render-queue monitor. Empty disables.
${user_config.RENDER_QUEUE_URL}HUGGINGFACE_TOKENHuggingFace access token for gated AI models (some Layer 9 skills need it). Create at https://huggingface.co/settings/tokens.
${user_config.HUGGINGFACE_TOKEN}OBS_WEBSOCKET_URLobs-websocket v5 endpoint for the obs-websocket skill and the live-ops agent.
${user_config.OBS_WEBSOCKET_URL}INCOMING_MEDIA_DIRDirectory the incoming-watch monitor will poll for new media (triggers auto-probe + QC gate). Empty disables the watcher.
${user_config.INCOMING_MEDIA_DIR}DEFAULT_VMAF_TARGETDefault VMAF score quality target for automated QC gating (0–100). 93 is broadcast-acceptable; 95+ is indistinguishable from source.
${user_config.DEFAULT_VMAF_TARGET}SHAKA_KEY_SERVER_URLDRM key-server URL for Shaka Packager (Widevine + PlayReady + FairPlay cbcs). Used by media-shaka and ffmpeg-drm skills.
${user_config.SHAKA_KEY_SERVER_URL}DEFAULT_ENCODE_PRESETDefault x264/x265 preset when the user doesn't specify one. Slower = better compression. Options: ultrafast, superfast, veryfast, faster, fast, medium, slow, slower, veryslow.
${user_config.DEFAULT_ENCODE_PRESET}OBS_WEBSOCKET_PASSWORDobs-websocket v5 password. Find it in OBS → Tools → obs-websocket Settings → Show Connect Info. Leave empty if auth is disabled.
${user_config.OBS_WEBSOCKET_PASSWORD}CLOUDFLARE_STREAM_TOKENCloudflare Stream API token for media-cloud-upload direct uploads.
${user_config.CLOUDFLARE_STREAM_TOKEN}SAFETY_REQUIRE_CONFIRM_OVERWRITEWhen true, the pre-ffmpeg-validate hook blocks ffmpeg commands that would overwrite an existing output path without -n or explicit confirmation. Turn off for batch automation.
${user_config.SAFETY_REQUIRE_CONFIRM_OVERWRITE}Process videos using VideoDB Python SDK. Upload, search, edit, generate subtitles, transcode, capture, and stream — all via natural language.
Video generation at scale. Generate videos, images, and audio with Runway's API — batch ad campaigns, product videos, multishot stories, and creative iteration. Supports seedance2, gen4.5, veo3, Nano, Banana Pro, and more.