MCP server category
Image, video and audio MCP servers: comparison and full list
The Image, video and audio category lists 67 open-source MCP servers, checked on GitHub. The best scored on our grid are OpenScreen, Remove AI Watermarks, and FunASR. 82% received a commit in the last three months.
What Image, video and audio MCP servers are for
Multimedia MCP servers let an AI agent work with images, videos, recordings and transcripts. They support content teams, editors and people who need to search audiovisual archives. Available functions vary across tools and may include recording search, local transcription or video editing controls. FunASR describes audio transcription performed on the device. Screenpipe supports searches across its local recordings, transcripts and computer activity. A practical request might be: “Find the discussion about the product launch in the available transcripts and identify the material I should review before editing.” You still need to check the source material and the resulting output.
Example uses:
- Transcribe a local audio file with FunASR to prepare notes from a recorded meeting.
- Find a discussion in the local transcripts and recordings made available through Screenpipe.
- Prepare an editing timeline containing clips, transitions and animated text through FableCut.
- Control an OpenScreen screen recording to document the steps in a product demonstration.
The category in figures
Of the 67 servers in the category (2% of the directory), 12 are official vendor implementations, 90% declare a licence, 12 offer a remote server and 18 need at least one API key or token according to the MCP registry. The median repository has 69 stars.
All Image, video and audio MCP servers
Click a header to sort. Score out of 100: adoption, maintenance and framework (licence, registry, official status); risk: system access, secrets, licence, abandonment.
| Server | Stars | Last commit | Licence | Risk | Score |
|---|---|---|---|---|---|
| Screenpipe Search your local screen recordings, audio transcripts, and computer activity from screenpipe. | 22,054 | Oct 24, 2026 | — | moderate | 75 |
| FunASR Transcribe local audio with FunASR and SenseVoice using private, on-device inference. | 20,804 | Oct 22, 2026 | MIT | low | 85 |
| Remove AI Watermarks Identify and remove AI watermarks and provenance metadata (Gemini and Veo marks, SynthID, C2PA, EXIF) in images and video you generated your | 6,115 | Oct 24, 2026 | Apache-2.0 | low | 88 |
| OpenScreen ◆ Free, open-source screen recorder and video editor that agents drive over stdio (the desktop app's own executable started with mcp): record | 4,867 | Oct 24, 2026 | MIT | low | 97 |
| OpenPost Manage social publishing workflows through OpenPost | 755 | Oct 23, 2026 | AGPL-3.0 | low | 74 |
| FableCut Browser video editor driven through a JSON timeline: clips, keyframes, transitions, text animations, filters, reference-video analysis and f | 746 | Oct 23, 2026 | MIT | low | 79 |
| Anki MCP Server MCP server for Anki flashcards: adaptive review, notes, media, and deck management via AnkiConnect. | 531 | Oct 22, 2026 | MIT | low | 78 |
| MCP youtube transcript An MCP server retrieving transcripts of YouTube videos | 492 | Oct 21, 2026 | MIT | low | 72 |
| MCP for Adobe Premiere Pro Local-first MCP for supported Adobe Premiere Pro workflows; start with a read-only connection check. | 433 | Oct 24, 2026 | MIT | low | 71 |
| Nanobanana MCP server An MCP server that provides image generation and editing capabilities | 401 | Oct 20, 2026 | MIT | moderate | 71 |
| Imagesorcery MCP Computer vision-based image recognition and editing tools. | 333 | May 19, 2026 | MIT | low | 58 |
| Gflow cli Drive Google Flow with your Google account: Veo text-to-video, image-to-video and clip extension, Imagen text-to-image, reusable characters | 332 | Oct 24, 2026 | MIT | low | 75 |
| Audio Edit, analyze and convert audio: loudness, spec checks, denoise, EQ, cuts, BPM, key. No ffmpeg. | 329 | Oct 21, 2026 | MIT | low | 70 |
| Guaardvark Self-hosted AI studio exposing local image, video and music-video generation, upscaling, RAG and code intelligence, with renders queued on y | 282 | Oct 24, 2026 | MIT | low | 75 |
| Audionaut Edit Audionaut multitrack projects: cut, fade, analyse, export. Requires the Audionaut app. | 265 | Oct 24, 2026 | — | moderate | 59 |
| WeftCut ◆ Drive the WeftCut desktop video editor: trim and split clips, keyframes, effects, captions, audio mixing, markers and export, with live edit | 261 | Oct 24, 2026 | MIT | low | 79 |
| BeatDesign Local-first AI media workbench with a canvas, short-form timeline editor and shared assets, controllable by agents. | 256 | Oct 21, 2026 | Apache-2.0 | low | 74 |
| MCP arr MCP server for *arr media suite - Sonarr, Radarr, Lidarr, Readarr, Prowlarr | 234 | Oct 16, 2026 | MIT | moderate | 69 |
| MCP video Local, guardrailed video editing: FFmpeg edits, captions, audio, effects, Hyperframes, repurposing, resumable workflows, quality gates and p | 220 | Oct 24, 2026 | Apache-2.0 | low | 79 |
| Sonicmatch Analyze video mood and pace, then recommend license-safe BGM with a mix spec. | 153 | Oct 23, 2026 | MIT | moderate | 67 |
| Ffmpeg MCP Local video search, trimming, stitching and playback through conversation, using the ffmpeg command line. | 153 | May 20, 2026 | MIT | low | 55 |
| Scumble Agents drive a layered AI inpainting editor: select by text, inpaint, mask, layers, PSD/ORA export. | 150 | Oct 24, 2026 | GPL-3.0 | low | 67 |
| ClipmivoAI AI video workflows: model discovery, credit quotes, generation, private media and usage. | 142 | Oct 15, 2026 | MIT | moderate | 73 |
| BibiGPT Summarize, transcribe and search YouTube, Bilibili, TikTok, podcasts and other videos with AI. | 131 | Oct 10, 2026 | — | moderate | 56 |
| Oh My Cassette ◆ Turn raw clips into a finished cut through chat. | 117 | Oct 22, 2026 | MIT | moderate | 81 |
| Reaper MCP (xDarkzx) Music production in REAPER: transport, tracks, MIDI editing, mixing and FX, tempo maps, markers and regions, and ReaScript execution. | 107 | Oct 23, 2026 | Apache-2.0 | low | 65 |
| MCP video analyzer Video analysis: extract transcripts, key frames, OCR text and annotated timelines from Loom links and direct MP4 or WebM files. | 102 | Oct 21, 2026 | MIT | low | 65 |
| Openrouter MCP multimodal Multimodal access to OpenRouter models: text chat, image, audio and video analysis, and image, audio and video generation (Veo, Sora, Seedan | 101 | Oct 24, 2026 | Apache-2.0 | moderate | 70 |
| Text from image Check from a screenshot that text is really visible to users. Local, offline OCR. | 94 | Oct 18, 2026 | — | moderate | 55 |
| LivePilot 317-tool agentic MCP production system for Ableton Live 12 — device atlas, sample engine, composer | 89 | Oct 22, 2026 | — | moderate | 54 |
| Music Studio ABC sheet music + Strudel live coding: songs, videos and live jam sessions in ext-apps widgets. | 75 | Oct 19, 2026 | AGPL-3.0 | low | 64 |
| Transloadit Media Processing Media processing via Transloadit Robots: video encoding (HLS, H.264, VP9), image resizing and watermarks, document conversion, OCR and trans | 73 | Oct 24, 2026 | MIT | moderate | 69 |
| Audio analyzer rs Gives LLMs ears. Spectral, harmonic, rhythm, stereo, and structural audio analysis. | 70 | Mar 15, 2026 | MIT | low | 43 |
| Suno by Ace Data Cloud Suno AI music generation, lyrics, covers and vocal extraction via the Ace Data Cloud API. | 69 | Oct 21, 2026 | MIT | moderate | 68 |
| Immich Photo Manager Manage a self-hosted Immich photo library: CLIP natural language search, GPS album curation, duplicate detection, library health reports and | 59 | Oct 6, 2026 | MIT | moderate | 68 |
| Ableton Live MCP Server Control Ableton Live with 156 typed tools for music production and offline .als analysis. | 58 | Oct 23, 2026 | MIT | low | 63 |
| Runapi ai MCP Unified API for AI models across providers: browse models and prices, create image, video, music and audio tasks, poll results and call LLM | 56 | Oct 23, 2026 | Apache-2.0 | moderate | 67 |
| Sonilo Music & Sound Effects Licensed soundtracks and sound effects matched to your video. Commercial use OK. | 52 | Oct 20, 2026 | MIT | moderate | 62 |
| Exif MCP Examine image metadata such as EXIF, XMP, JFIF and GPS for searching and analyzing photo libraries. | 40 | Nov 14, 2025 | BSD-2-Clause | low | 41 |
| WaveSpeedAI MCP ◆ Run image, video, audio and 3D models from the WaveSpeed catalog: model search, schema introspection, local file upload and price quotes bef | 34 | Oct 21, 2026 | MIT | moderate | 75 |
| Talkthrough Turn screen recordings or other local video and audio into local Whisper transcripts, keyframes, OCR and full-text search, with wall-clock t | 32 | Oct 24, 2026 | MIT | low | 65 |
| Video edit MCP Video and audio editing: trim, merge, effects, overlays, format conversion, audio processing and YouTube downloads, chaining steps without i | 30 | Jul 30, 2025 | MIT | moderate | 32 |
| Prompt to asset Generate app icons, favicons, OG images, logos and wordmarks across many image models, validating contrast, text accuracy and palette before | 24 | Oct 18, 2026 | MIT | low | 64 |
| After effects MCP (a-y-ibrahim) Control Adobe After Effects: run ExtendScript, render in the background and inspect comps and layers, with Arabic/RTL and multilingual suppo | 22 | Oct 10, 2026 | MIT | low | 58 |
| Clipkit ◆ Compose motion-graphics videos from a single JSON document: author, validate and preview frames, add captions and promo templates, and rende | 19 | Aug 12, 2026 | Apache-2.0 | low | 68 |
| Aidemo Turn a storyboard JSON into a narrated, captioned product-demo MP4 (TTS voiceover, Whisper captions, auto-zoom) via Chrome replay and ffmpeg | 18 | Oct 21, 2026 | MIT | low | 63 |
| Gif creator MCP Create GIFs from videos. | 18 | Jun 9, 2025 | MIT | moderate | 30 |
| Torrentclaw MCP ◆ Search and discover movies and TV shows with torrent links, quality scoring, streaming availability and cast/crew metadata. | 16 | Sep 7, 2026 | MIT | low | 62 |
| Editmamei Natural-language interface for Adobe Photoshop that plans edits and executes them with native Photoshop tools rather than generative AI. | 15 | Oct 24, 2026 | — | moderate | 52 |
| Cochlea Offline, deterministic audio engine: compose scores as data, render WAV or FLAC, report loudness, pitch, tempo and key, generate spectrogram | 15 | Sep 19, 2026 | Apache-2.0 | low | 57 |
| Sonos ts MCP Sonos control via UPnP: device discovery, multi-room playback, queues, music library browsing, alarms, event subscriptions and EQ settings. | 15 | Nov 27, 2025 | — | moderate | 27 |
| Image to SVG Convert raster images (PNG, JPG, WebP, TIFF) to SVG from base64 or URL input, locally or via Docker. | 10 | Jul 9, 2026 | Apache-2.0 | low | 48 |
| Gemini Media Gemini media generation: Nano Banana images and editing, Veo video, text-to-speech and Lyria music, via the Gemini API or Vertex AI. | 10 | Oct 18, 2026 | Apache-2.0 | low | 60 |
| Dali by Lulu Score AI image and video generation prompts for models like Veo 3, Sora, Kling, Midjourney and Flux, and rewrite weak ones before generating | 9 | Jul 13, 2026 | MIT | low | 48 |
| Media gen MCP Media generation, editing and asset downloads with OpenAI Images/Videos and Google GenAI (Veo). | 9 | Jan 30, 2026 | MIT | moderate | 40 |
| Macos vision MCP Offline OCR and image analysis via Apple's Vision framework: text from images and PDFs with bounding boxes, face, barcode and QR detection, | 7 | Aug 25, 2026 | MIT | low | 53 |
| Topaz MCP AI image enhancement (upscaling, denoising, sharpening) via the Topaz Labs API, with models such as Standard V2, Wonder 2, Bloom and Recover | 7 | Feb 16, 2026 | MIT | low | 33 |
| Aetherwave MCP Music, image, video and audio generation across Suno, Grok Imagine, Seedance, Kling, Veo, Ideogram and more: generate, edit, upscale, refram | 6 | Oct 5, 2026 | MIT | moderate | 58 |
| Autoeq MCP AutoEQ headphone and IEM EQ data: search by name or sound signature, get parametric EQ settings, compare models band by band and browse Harm | 6 | Mar 21, 2026 | MIT | low | 33 |
| Trident (Tripo 3D) AI 3D model generation via Tripo (text, image or multiview to 3D) plus retopology, format conversion (GLB, FBX, OBJ, STL, USDZ) and stylizat | 5 | Oct 18, 2026 | Apache-2.0 | low | 57 |
| Clipy MCP Read-only access to Clipy screen recordings: search the library, fetch timestamped transcripts, AI summaries and key moments with video fram | 5 | Oct 12, 2026 | MIT | moderate | 63 |
| Cleanor ◆ Image optimization and conversion, SVG QR codes, Cleanor Labs storage and image-format data, and dev utilities such as hashes, UUIDs, JWT de | 1 | Oct 7, 2026 | MIT | low | 60 |
| Rendobar ◆ Submit media-processing jobs to the Rendobar API with local file upload: transcode, run raw FFmpeg, burn in captions and add watermarks. | 1 | Oct 22, 2026 | MIT | moderate | 60 |
| Spicy MCP ◆ Official SpicyAPI server: browse 200+ image, video, audio and chat models (Seedance, Kling, Wan, MiniMax, Suno), get exact USD quotes, uploa | 1 | Oct 23, 2026 | MIT | low | 55 |
| Fast Video Cataloger ◆ Search, view and tag a local video library cataloged by Fast Video Cataloger: transcripts, scene thumbnails, visual search, keywords and bin | 0 | Oct 7, 2026 | MIT | moderate | 60 |
| VidWords YouTube ◆ Search YouTube video transcripts and frames (slides, charts, on-screen text) and get matching moments with timestamped links and verified ci | 0 | Oct 21, 2026 | MIT | low | 60 |
| AppsGolem YouTube Cutter ◆ Cut clips from YouTube videos via the AppsGolem API in a chosen format and quality, and get a direct download URL. | 0 | Sep 4, 2026 | MIT | moderate | 55 |
Official servers
Published by the service's own vendor (per the awesome-mcp-servers list): they usually track the API closely and are better maintained.
- OpenScreen 4.9K ★
- WeftCut 261 ★
- Oh My Cassette 117 ★
- WaveSpeedAI MCP 34 ★
- Clipkit 19 ★
- Torrentclaw MCP 16 ★
- Cleanor 1 ★
- Rendobar 1 ★
- Spicy MCP 1 ★
- Fast Video Cataloger 0 ★
- VidWords YouTube 0 ★
- AppsGolem YouTube Cutter 0 ★
Choosing and securing
Start with supported formats and documented operations, such as transcription, search, recording or editing. Check whether processing happens on your device or through a remote service. Review repository maintenance, licensing, dependencies and file permissions. For editing workflows, look for preview and recovery options. For transcription, check supported languages and the stated hardware requirements before selecting a server for your workflow.
Recordings and screenshots can reveal conversations, faces, private screens and visible credentials. Check which files the agent can access and where uploads are sent. Preserve originals before editing or deleting media. Audio and visual content can also carry instructions intended to redirect the agent. Review usage rights and retain provenance information that helps identify the origin and editing history of media.
Before connecting a server to an agent with access to your data, run the AI agent security check.
Most recent
- Fast Video Cataloger created October 7, 2026
- Spicy MCP created October 5, 2026
- ClipmivoAI created September 14, 2026
- Sonicmatch created September 10, 2026
- Scumble created September 7, 2026
Other categories
Learn more
Frequently asked questions
What is the best Image, video and audio MCP server?
On our grid (adoption, maintenance, framework), OpenScreen (97/100), Remove AI Watermarks (88/100), and FunASR (85/100) come first. The right choice mostly depends on the service you need and the rights you are willing to grant.
Are these servers free?
90% declare an open-source licence and are free to use. 18 need an API key: the connected service may be paid.
Can my audio files stay on my computer?
Some tools describe local execution, including FunASR for transcription. That does not mean the entire workflow stays local: the client or model may receive the resulting text. Check how files, transcripts and logs move through the system before processing recordings that contain private or confidential information.
Can I request a complete video edit through a conversation?
That depends on the operations exposed by the server. FableCut, for example, describes a JSON timeline with several editing elements. Provide the source media and clear instructions, then inspect a preview of the result. The final output also depends on the available footage and the settings you choose.
Related articles
- Build your own MCP server in 30 minutesOctober 24, 2026
- Automate your inbox with an AI agentOctober 18, 2026
- MCP, A2A, and tool calling: agent protocols explainedOctober 17, 2026
- Build an AI agent with n8n, without codingOctober 16, 2026
Neutrality. No affiliate links. No vendor pays to be reviewed, rated or ranked. Reviews are directed by François Chollet, neutral and factual, and every data point is dated and sourced. How we score