OneTool is a unified MCP server that exposes over 100 tools through the Python API, significantly reducing token consumption and cost when AI calls tools, avoiding context pollution, and supporting multiple functions such as web search, database, file operations, and chart drawing.

MCPollinations is a multimodal AI service based on the Model Context Protocol (MCP) that supports generating images, text, and audio through the Pollinations API. It offers a lightweight service that does not require authentication, is compatible with multiple AI models, and supports image saving and Base64 encoding return.

This is an Evernote integration server based on the Model Context Protocol, providing functions such as note management, notebook organization, tag system, and advanced search. It supports automatic conversion between Markdown and Evernote formats and includes OAuth authentication and change polling mechanisms.

Cronicorn is an intelligent HTTP job scheduling platform that automatically analyzes HTTP response content through AI, dynamically adjusts polling frequencies, coordinates multi - endpoint tasks, and implements automatic error recovery. Users only need to describe business logic in natural language.

A multimodal MCP server based on the Pollinations API, supporting the generation of image, text, and audio content

A server for the Pollinations.ai service interface based on the MCP protocol, providing image and text generation functions

This project demonstrates MCP poisoning attacks against AI workflows, including verification of multiple attack scenarios such as code generation pollution and financial data leakage, for educational research purposes.

An experimental chat project based on the MCP protocol, implementing long polling chat functions in a two-person room by abusing the MCP tool call mechanism.

Telegram MCP Server is a tool for remotely controlling AI programming assistants (Claude Code/Codex) via Telegram. It supports multi - session management, file operations, and smart polling for a truly unattended programming experience.

A MCP server for remote control of Claude Code and multi - agent team management based on Slack. It supports sending commands, receiving results, and coordinating multi - agent teams through Slack channels, and provides advanced functions such as persistent state, background polling, and universal approval hooks.

A Tavily MCP server that supports multi-API key load balancing, provides both SSE and HTTP interfaces, and can automatically poll keys to improve concurrency and availability.

An RSS MCP server based on TypeScript that supports parsing standard RSS/Atom feeds and RSSHub feeds, providing multi - instance polling and structured content output.

An MCP proxy server that filters tools from the upstream MCP server through a regular expression deny list, solves the context pollution problem, and reduces useless tokens

The Napkin AI MCP Server is an unofficial community - maintained tool that provides AI assistants (such as Claude) with the ability to generate visual content such as infographics, mind maps, and flowcharts through the Model Context Protocol. It supports multiple output formats (SVG, PNG, PPT) and storage backends (local, S3, Google Drive, etc.), and includes asynchronous processing, automatic polling, and retry mechanisms.

Flux ImageGen MCP Server is a dedicated server based on the Model Context Protocol (MCP), focusing on providing image generation and processing functions through Pollinations AI. It supports generating image URLs from text prompts, directly generating images, and managing multiple image generation models.

A tool for calculating the altitude, azimuth, rise, and set times of celestial bodies (sun, moon, planets, stars, and deep - sky objects) at any location on Earth, supporting light - pollution analysis.

This project is an MCP server based on Node.js and Express.js, used to interact with the OpenWeather API to obtain air pollution data based on latitude and longitude or city and country, and implement secure API key management.