Playwright MCP is a browser automation server based on Playwright that enables interaction between LLM and web pages through structured data rather than pixel input.

Pixelle MCP is an omnimodal AIGC solution based on the MCP protocol, which can convert ComfyUI workflows into MCP tools without code, achieving seamless integration between LLM and ComfyUI.

Playwright MCP is a browser automation service based on Playwright. It provides web page interaction capabilities for LLM through structured accessibility snapshots rather than pixel input. It supports multiple functions such as headless/headed mode, tab management, and navigation operations, and is suitable for scenarios such as web page navigation, data extraction, and automated testing.

Pixeltable's multi-modal model context protocol server set provides indexing and querying functions for audio, video, images, and documents

A web screenshot tool optimized for AI vision workflows. Automatically segments the entire web page into 1072x1072 pixel tiles to ensure the best processing effect. Supports MCP integration and screen recording functions

A web page screenshot tool optimized for AI vision workflows, automatically splitting into 1072x1072 pixels to meet the processing requirements of the Claude Vision API

LA FORGE MCP is a Model Context Protocol server that provides visual perception capabilities for AI programming assistants. Through pixel-level difference comparison, computed style extraction, and CSS rule chain analysis, it helps AI accurately detect CSS/HTML rendering issues, addressing the pain point that AI can only read code but cannot truly 'see' the rendering results.

The Agent skill pack of the SpriteCook AI game asset generation tool, used to guide AI coding assistants to generate pixel art and HD game materials through the MCP tool.

AnyPixel.js is an open - source hardware and software library for creating large - scale, non - traditional, interactive display devices. It provides a complete hardware design, firmware, back - end, and front - end framework, supporting developers to build web - based interactive pixel display systems.

This project provides a cloud-based serverless Playwright browser service through Playwright-MCP. It uses Playwright's accessibility tree instead of pixel input to achieve fast and lightweight web page automation operations. It supports structured data interactions and is suitable for scenarios such as web page navigation, form filling, data extraction, and automated testing.

This project is a collection of multimodal model context protocol servers by Pixeltable, providing indexing and querying services for audio, video, images, and documents, and supporting local Docker deployment.