Image recognition MCP servers

A server here analyses or describes what is in an image or video: captions, object detection, visual question answering.

Not here: OCR of documents, generating images and editing images.

236servers
112 remote124 localTagged by Odel's AI, which can make mistakes

Best fit

135 made for this, ranked by starsRemoteLocal
  1. Klaket @huseyinstif/klaket-mcpLet AI agents watch videos: local transcripts, speakers, scenes, chapters and moment search6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.3 GitHub stars.
  2. Video URL Analyzer @u2n4/video-url-analyzer-mcpAnalyze YouTube/TikTok/Instagram videos: transcripts, AI insights, tutorial extraction6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.3 GitHub stars.
  3. Vidin @teunlao/vidinTurn a screen recording into agent-readable annotated keyframes. MCP server + CLI on ffmpeg.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.2 GitHub stars.
  4. Image Reader MCP @sylphxai/image-reader-mcpEvidence-first image MCP. Agent Media Twin with metadata and citeable OCR evidence.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.2 GitHub stars.
  5. Media Context @vishalguptax/media-context-mcpLocal MCP server to analyze video, audio & images: frames, transcripts, on-screen text.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.2 GitHub stars.
  6. GPU Bridge @fjnunezp75/gpu-bridge30 GPU-powered AI services as MCP tools. LLM, image, video, audio, embeddings & more.9% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.2 GitHub stars.
  7. ReftrixMCP @tkmd/reftrixmcpWeb design analysis with 26 MCP tools: layout, motion, quality, semantic search via pgvector5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.2 GitHub stars.
  8. Toolpiper @modelpiper/toolpiperOn-device MCP tools for macOS: local AI, speech, vision, browser automation, system control.10% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from a package registry.2 GitHub stars.
  9. FlowScope UX Flow Analyser @davidngugi/flowscopeTurn YouTube product demos into structured UX flow reports, and compare demos side by side.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1 GitHub stars.
  10. PIL @pjpoulose/pilPIL: local-first knowledge base for your Instagram saved posts, with a read-only MCP server.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1 GitHub stars.
  11. Visiongemma @q1023884985/visiongemmaLocal AI vision MCP server for Windows: OCR, chart analysis, scene description. 100% offline.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1 GitHub stars.
  12. Kiyas @saiffmirza/kiyasAI design fidelity — compare Figma designs or screenshots against rendered UI. No API keys.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
  13. OcularAudio @rayakaan/ocular-audio-mcpMCP server for video transcripts, screenshots, and OCR on YouTube and web videos.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
  14. Yolo @rjn32s/mcp-yoloAn MCP server providing zero-shot object detection and segmentation using Ultralytics YOLOE.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1 GitHub stars.
  15. NVIDIA NIM @david-eve-za/nvidia-nim-mcpMCP server for NVIDIA NIM - 50+ LLMs, multimodal, image gen, embeddings, reranking5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
  16. Flickr @mk3core/flickr-mcpFlickr MCP server — tag photos, write descriptions, set titles, and add photos to groups using AI.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
  17. BEHAVR @confuse57/behavr-mcpMCP server for BEHAVR: remember images/videos and ask questions about them5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
  18. Fast Video Cataloger @videocataloger/fvc-mcpSearch, view and tag your local video library from Claude: transcripts, scenes, visual search.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.
  19. TinkyVision @agewellepm/tinkyvisionScreen observation, OCR and permission-controlled app input over MCP for Apple silicon Macs.10% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.
  20. Ollos @kelvinbiffi/ollos-mcpLocal, offline transcription, speakers, keyframes, on-screen text and review of any audio or video.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
  21. AltText @alttextai/alttext-mcpGenerate alt text, translate descriptions, and manage your AltText.ai image library.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
  22. AdTest @adtest/adtest-mcpAI-only ad scoring for Claude: score image, video or text ads on 13 dimensions before you spend.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
  23. Secret MCP @yyeongjin/secret-mcpLocal MCP server for isolated, evidence-backed GDWEB design analysis and web search.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
  24. ocular @xyun1996/ocularVision tools for coding agents: screenshots, OCR, UI diffs, errors, tables, and charts.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.