Claude Desktop Руководство установки
Десктопный клиент с поддержкой локальной конфигурации MCP.
MCP Servers

AI-driven React UI effects library — 64 composable animations (cursor, scroll, text, background). Hooks + CSS + Components. CLI & MCP powered.

This is an MCP Server that wraps various JVM diagnostic tools (CLI) such as jstat, jcmd, jps, adds a nice prompt to improve how they work, and exposes them.

An MCP server to use with iOS and Android. Seamlessly maps Native screens and WebViews. Multiple parallel devices supported.

This project provides a modular, resource-oriented FastMCP server for interacting with Microsoft Graph API.

Give Claude full access to your email through Mozilla Thunderbird. CLI + MCP server for AI agents.

A terminal AI agent you can fully undo - every file and shell action is snapshotted and reversible. Model-agnostic (any LLM), and connects to 1000+ app tools and MCP servers.

mTarsier - The Open Source MCP & Skill Manager for Claude, Cursor, VS Code & any AI client.

Your AI agent doesn't know what's broken in production. This fixes that.

- Access MATLAB documentation via matlab://documentation/getting-started URI - Get started guide with examples and usage instructions

An MCP server that gives AI assistants direct access to your Help Scout inboxes, conversations, customers, organizations, threads, and Docs knowledge base.

This is a server that supports remote MCP connections, with Google OAuth built-in and provides an interface to the Google Tag Manager API.

Live through COM on Windows, or headless through ezdxf anywhere — one typed contract, two engines.
Agent Skills

微信读书前几天上了一个官方 AI skill(weread.qq.com/r/weread-skills),把书架、笔记、阅读统计、推荐 6 件事开放给 AI。能力很强,但它只是「自然语言包装的搜索接口」——你让它推荐书,它不会去看你已经读过什么,常把你笔记 68 条的书又当新书推回来给你。

For engineers who want agents shipping code without a human in the loop.

A hypothesis-driven debugging skill for Claude Code, Codex, and Gemini CLI. Instead of blindly guessing fixes, it instruments your code with runtime logs, generates multiple hypotheses, and...

Environment-Runtime 是一个面向工程场景的轻量级 Agent Runtime。随着 OpenClaw 在企业工程流程中的关注度持续提升,一个现实需求也逐渐显现:面向通用场景的 agent runtime 设计未必能直接覆盖所有工程落地形态。这个项目把视角收窄到高确定性工程流程,希望把常见工程任务组织成更清晰、更可复用、更容易维护的运行时。

Human-defined workflows. AI-powered execution. YAML decision trees with scheduling, hooks, MCP, and 21 built-in tools.

There are plenty of skills out there. This library avoids slop and only ships skills that:

A lightweight .NET AI SDK for provider routing, streaming, tool calling, and chat session management.

One-page team capacity: load, buffer, explicit cuts — honest ranges.

Azur Lane Resources extracted and parsed by Fernando2603/Azurex

Per-task rubric-blind skills for the Parthenon self-evolving legal-agent framework (Harvey LAB).

PrediHermes is a Hermes skill for geopolitical market forecasting. It wires together:

A Claude Code skill that rewrites text into plain language per ISO 24495-1:2023 — the international standard that defines plain language by reader outcome: the intended readers can what they need, it...