MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
MCP server for human-in-the-loop image annotation: the user draws a box / point / polygon on a screenshot, the agent gets pixel coordinates, a name and a cropped template image.
MIT ★ 1 -
annomate-mcpgithub.com/caliperhq/annomateupdated this year
MCP server for VIA v3 — lets Claude read and write image annotations in real time
pypi MIT License Copyright (c) 2026 Mike Doty Permi -
Self-hosted data annotation and labeling platform for images, video, audio, text, PDF and LLM evaluation. Your storage, your models, your cloud.
pypi NOASSERTION -
Local-first image dataset curation studio for inspection, annotation, OCR, PII detection, embeddings, segmentation, COCO, YOLO, exports, MCP, and modular AI plugins.
MIT ★ 2 -
mcp-screenshot-servergithub.com/aamar-shahzad/mcp-screenshot-serverupdated this year
MCP server for screenshot capture and image annotation with boxes, lines, arrows, and text
pypi MIT -
markitup-mcp-servergithub.com/smythmyke/markitup-mcp-serverupdated this year
MCP server for MarkItUp — AI image annotation and marketing visuals via Claude + Gemini
npm MIT -
Image Annotation, including Histopathology slides; From H&E slide to training data in one conversation: nuclei, outlines and slide QC, exported to QuPath, COCO or YOLO. No GPU needed. For Whole Slide Imaging (WSI), Digital and Computational Pathology, Bioinformatics, Tumor Microenvironment (TME) and MLOps Data Labeling.
pypi MIT -
GroupDocs.Annotation for .NET MCP server — add, list, and remove annotations on PDF, Word, Excel, PowerPoint, and image documents via AI agents.
nuget MIT -
covasgithub.com/sevenking-dev/covasupdated this year
Single-image annotation workspace — draw boxes, arrows, freehand marks, send annotated results back into Codex conversations.
MIT -
MCP server that gives AI agents read-only access to whole-slide images and QuPath annotations
Apache-2.0 -
Stdio bridge to the local MCP server of Snipaya, the macOS image annotator
npm MIT -
pdf-reader-mcpgithub.com/Xvvln/pdf-reader-mcpupdated this year
MCP server for extracting text, images, tables, links, annotations, and metadata from PDF files
pypi MIT ★ 1 -
GroupDocs.Redaction for .NET MCP server — redact text, image regions, annotations, and metadata in PDF, Word, Excel, PowerPoint, and image documents via AI agents.
nuget MIT -
crawsgithub.com/Aver005/crawsupdated this year
🦀 Craws — a BLAZING-fast, automation-first image editor built for developers. Powered by a tiled, content-hash-cached Rust engine that works in linear-light f32 color, it ships as a headless CLI and an MCP server so AI agents can annotate, compose, and transform images directly. 🖼️⚡ Pixels never travel as JSON — every claim is backed by benchmark
-
mcp-server-roboflowgithub.com/mayankd409/roboflow-mcp-serverupdated this year
Hardened MCP server for the Roboflow API: dataset ingestion, image tagging, annotation upload, version lifecycle. Capability-scoped, audited, SSRF-guarded.
pypi Apache Software License -
An offline PDF engine you can use yourself or hand to an agent: a native Windows viewer + page editor — split, splice, annotate, fill forms, sign with an image, redact — plus a cross-platform 19-tool MCP server for filling, annotating, and redacting PDFs with any MCP-compatible client. No telemetry, no network. PySide6 · PyMuPDF. AGPL-3.0-or-later.
pypi NOASSERTION ★ 3 -
MarkItUp MCP Serverno update data
MCP server for MarkItUp (https://markitup.app) — AI-powered image annotation and marketing-visual generation. Lets Claude Code, Claude Desktop, Cursor, ChatGPT-with-MCP, and other MCP-compatible clients call MarkItUp directly. Powered by Claude analysis + Gemini image generation. ## Tools (5) - **markitup_credit_balance** — check credits + subscription status (free) - **markitup_generate** — create 3 marketing-visual variations from any image (1 credit) - **markitup_regen** — regenerate one variation from a previous call (1 credit) - **markitup_extend** — AI outpainting for aspect-ratio changes (1 credit) - **markitup_remove_background** — HD background removal via Photoroom (1 credit) Get your API key at https://markitup.app/dashboard/.
smithery -
Annotate screenshots for user guides. Saved as plain JPEG/PNG whose boxes and notes stay editable. Web, Chrome extension and MCP.
npm MIT ★ 1 -
Document Suite: open-source PDF and image viewer and editor for React, Vue and Angular, with a REST API and an MCP server. Self-hosted and MIT licensed.
npm MIT -
🖌️ An image markup & drawing tool on one shared C++ core, scriptable in its own .stc language: a Qt desktop app, a browser app compiled to WebAssembly, and a Zig CLI, plus a Chrome extension, a VS Code extension for .stc, a Rust MCP server, a Go server, a Python package, and a dotnet bot.
Apache-2.0 -
DICOM ROI annotation & analysis desktop tool (PySide6): SAM / MedSAM / SAM-Med2D-3D interactive segmentation + 5 classical CV ROI detectors, with an MCP bridge for AI-agent driven workflows. 医学影像 ROI 标注与分析桌面工具。
-
Preservation-aware visual dataset curation & contrastive safety alignment engine for AI. Replaces destructive whole-image sample-dropping and diffusion inpainting drift with deterministic sub-pixel Rust SDF masking, 540p JPEG survivability verification, and zero-glue MCP server for foundation VLM pre-training and DPO.
NOASSERTION ★ 1 -
vision-primitives-mcpgithub.com/zouyuanqing/vision-primitives-mcpupdated this year
Vision-primitives MCP server for text-only LLMs: describe, locate (coordinates), OCR with bbox, annotate, crop, zoom, automated anomaly scanning, computer use — powered by Xiaomi MiMo V2.5. Single-file Python, Pillow-only dependency.
MIT ★ 9