"Top 20 most starred MCP servers on GitHub" matching MCP servers:

Enables assistants to publish readable replies and status updates to e-ink displays and Home Assistant, with switchable characters and state retrieval through MCP tools.

Official MCP server for Theta EdgeCloud's On-Demand Model APIs, providing access to 20+ AI models including image generation, audio transcription, and LLMs, directly from MCP-compatible clients.

A FastMCP server that provides virtual try-on functionality through the Replicate API, allowing users to visualize how clothing items would look on models.

Give your AI assistant eyes and ears — analyze any video, audio, or image, entirely on your machine.

Enables AI agents to inspect and work with PDF files entirely on the local machine, covering page counts, metadata, layout, encryption permissions, scan-vs-text detection, and page rendering. Generated results land in a temporary workspace as short artifact ids that can be chained between tools and exported to real folders on request.

Enables AI agents to read public Instagram Reels without logging in, returning details, on-device timestamped speech-to-text transcripts, real video frames, preview comments, and Reel links harvested from web pages.

Compress, merge, split, OCR and convert PDFs (Word, Office, images, webpages) on local files, plus password protect and unlock, through the Papyrio API. Works without a key on a free guest tier; an optional RapidAPI key raises the limit.

Post AI-generated images to Vynly, the AI-only social feed. Four tools for posting, reading, searching, and ephemeral sparks — auto-claims a demo token on first run so it works with no signup.

Lets coding agents such as Claude Code and Codex hand summarising, field extraction, classification and OCR to the free, offline on-device model for Apple Foundation Models on macOS 27. One Rust binary, installed with Homebrew.

Zero-auth hosted MCP for AI builders: optimize/convert images (WebP/AVIF/JPEG), paste-ready SVG QR codes, plus cited Cleanor Labs data on device storage capacity and next-gen image-format savings. No API key or signup.

Local MCP server for PDF and photo tasks: merge, split, rotate and number PDFs, convert images to PDF, and inspect or strip hidden photo metadata (EXIF/GPS). Runs entirely on your machine — no uploads, no accounts, no API keys.

Enables inspecting, creating and modifying MakeHuman characters inside Blender — phenotypes, rigs, assets, skins, faces, expressions and saved presets — through 34 named tools served by a companion Blender bridge add-on that dispatches a closed command vocabulary instead of executing code. It also exposes read-only introspection, catalogue and documentation lookups that always answer, even when Blender is unreachable.

Enables Claude Code to generate and edit images using Google's Gemini and Imagen models on Vertex AI, with support for multiple models, aspect ratios, and image fusion.

Lets a text-only LLM read images and PDFs by delegating to a vision model running on your own GPU.

Enables Claude to drive Google Flow end-to-end: batch video generation on Veo 3 / Omni 1.1, scene chaining, keyframe transitions, video-to-video edits, and automatic downloads, all through a local bridge with the OmniFlow Chrome extension.

Turn a sentence into an image or a short video on your machine.

Enables precise extraction of PDF content, including text coordinates, vector shapes, images, and on-demand raster rendering, for use cases like reconstructing design files.

MCP server for screen recording on macOS, iOS Simulator, and Android, with key-frame extraction via ffmpeg, enabling AI agents to capture UI motion and transient visual states.

Enables an AI assistant to conversationally control Autodesk AutoCAD, drawing, inspecting, and editing plans, and includes a cross-platform DXF engine that runs without AutoCAD on macOS and Linux. It exposes tools for rendering views as images, batching geometry, blocks, and structures, measuring and checking plans, and undoing whole batches.

Search a YouTube video's transcript and get back only the moments that answer the question, each with a clickable timestamp deep link. Also reads the video's frames — slides, charts, demos and on-screen text — and verifies every citation against recorded evidence before returning it

Gives AI agents eyes on Hyprland Wayland desktops by capturing screenshots via grim and offering optional image analysis with Gemini API.

MCP server for 2DAI that enables AI assistants to generate images and videos, upload references, browse the public feed, and manage cloud drive on a 2DAI account.

Enables printing labels on King Jim Tepra PRO SR5900P over Wi-Fi/LAN, supporting Japanese text, QR codes, and images through natural language.

Creates narrated 9:16 documentary videos from user-described topics using Wikimedia images and AI voiceover, rendered entirely on your machine.

Connects an AI assistant to Instagram Edits on macOS through its native Automation endpoint, exposing discovered editing tools with an allowlist and explicit write gate so users can import media, adjust timelines, and export video.

Enables an assistant to watch a cooking video once and then guide a hands-free cook through it: importing videos into validated step-by-step guides, navigating steps, searching timestamped transcript and on-screen evidence, showing exact frames and clip replays, and setting persistent timers across restarts. Sessions, progress, and deadlines are stored so users can resume where they left off.

MCP server that exposes OpenAI image generation and editing as tools, focused on creating consistent character expressions by compositing edited regions onto a stable base image.

Enables engraving LilyPond sources into cropped, self-contained EPS, PDF, SVG, and PNG assets for placement in page-layout software, with support for files or inline code and helpful diagnostics on failure.

Enables AI clients to control a running Blender instance through the Model Context Protocol, allowing scene inspection, object creation and editing, rendering, and execution of Blender Python. It works locally via a WebSocket bridge between the MCP server and a Blender add-on.

This MCP server searches Google's public ads database and uses AI to analyze ad images/videos, with intelligent caching for performance.