Claude Desktop 설치 가이드
로컬 MCP 설정을 지원하는 데스크톱 클라이언트.
MCP Servers

这是一个基于大语言模型和MCP(Model-Controller-Plugin)和Rag架构的网络安全智能助手项目。它旨在通过自然语言交互,帮助用户执行渗透测试任务、查询安全信息、分析流量包等。

OnebotV11 Bot as MCP Server.

A Telnet-based MCP tool for Cisco network device CLI, optimized for Cisco IOS devices.

This MCP server uses SSE transport and is authenticated with an API key.

This is an unofficial, community-built project. It is not affiliated with, endorsed by, or maintained by HackerOne. "HackerOne" is a trademark of HackerOne, Inc.

mcp-repl is an MCP server that gives an agent a persistent Python or R session, kept alive across tool calls.

AI audio engineering system: MCP server + Claude Code skills for professional automated mixing and mastering

comprehensive Discord selfbot MCP server using discord.py-self

Prompt-based frameworks hope the LLM follows instructions. This one blocks the call if it doesn't.

A proof-of-concept MCP integration that controls a beacon by exposing AdaptixC2 APIs to gemini-cli

An MCP server for interacting with Linear's API. This server provides a set of tools for managing Linear issues, projects, and teams through Cline.

AI-powered browser automation for Claude Code by DebugBase
Agent Skills

每天定时 12:00 更新,数据从 2024-01-01 开始拉取,每天累计更新。

Feed your AI something healthier than Markdown. allium-lang.org

给 Agent 一个链接,它会先选对读取通道,再把完整原文交给后续的总结、写作、翻译或分析任务。

See the Kocoro product site for the current product demo.

git-memento is a Git extension that records the AI coding session used to produce a commit. It attaches AI conversation transcripts as git notes, creating an audit trail for AI-assisted development.

A Python CLI tool for coordinating multi-agent development workflows through shared JSON task files. Designed for use with OpenClaw and AI agent orchestration systems.

Central repository for managing Skills to "human" vibe-talking.

No network, no API key, no dependencies. One small C program.

https://github.com/user-attachments/assets/2a5760fe-1886-4490-9c2b-aceec24aff2f

Defense-in-depth protection for Claude Code. Blocks dangerous commands and protects sensitive files via PreToolUse hooks.

An OpenClaw-native knowledge retention skill that turns raw inputs into structured practice so you can use what you know, not just store it.

Productized companion to our paper (arXiv:2606.32014). See Paper & citation.