Cursor Руководство установки
AI-редактор кода, часто используемый с developer MCP.
MCP Servers
AgentKit: Build multi-agent networks in TypeScript with deterministic routing and rich tooling via MCP.
JSReverser-MCP 是一个面向 JavaScript 逆向分析的 MCP 工具，专门用于帮助开发者在真实浏览器环境中高效定位前端核心逻辑。它 将脚本检索、断点调试、函数 Hook、网络请求追踪、调用链分析、混淆还原和风险评估整合为统一能力，可直接接入 Claude、 Codex、Cursor 等支持 MCP 的客户端。
Give your AI agent eyes for PDFs — structured text, tables, OCR, visual evidence, and page-level citations via MCP. Native Rust, local-first.
Query MCP enables end-to-end management of Supabase via chat interface: read & write query executions, management API support, automatic migration versioning, access to logs and much more.
CodeGraph transforms your entire codebase into a semantically searchable knowledge graph that AI agents can actually reason about—not just grep through.
Connect ClickHouse to your AI assistants.
MCP server that gives AI assistants the power to control a web browser.
Browse the web, directly from Cursor etc.
Autonomous bug-bounty framework for Claude Code and 6 other AI coding tools — 50 agents, 26 commands, 19 CLI tools, 11 skills, 2 MCP servers.
Headless IDA Pro MCP Server
MCP Server for Turkish & American Stock Exchange and Fund Data
This repo contains the source code for running an MCP server that interacts with the Google Ads API.
Agent Skills
Uses Apple Reminders as a bridge between Claude on your phone and Claude Code on your Mac, dispatching agents from new reminders.
A compact high-signal benchmark for evaluating frontier agents
Market Map agent eval suite for the Reforge AI Evaluation course
A Claude Code skill that turns peer review reports into structured revision roadmaps. Extracts every reviewer request as a separate task, maps dependencies as a directed acyclic graph, validates the...
Publish Markdown or HTML articles to WeChat Official Account drafts with one command. Say goodbye to tedious copy-paste-format workflows.
"Other agents start cold. Yansu.skill starts warm."
把参考视频拆成可控制、可替换的视频结构，再按用户目标生成可直接使用的视频提示词。
An skill to plan, design, and build video-driven parallax hero landing pages using video scrubbing.
two intents (improve my prompt / build this whole thing), specifiability routing, One-Shot build prompts, team execution seeded from a build brief — digest
The Definitive AI Agent Benchmark Standardized, reproducible evaluation across 314 tasks, 33 domains, and 4 difficulty levels.
Skillify watches what you did in a conversation — the tools you used, the corrections you made, the workflow you followed — and captures it as a portable agentskills.io SKILL.md file.
AI agents that triage infrastructure alerts, investigate root causes, and propose fixes — while a solo operator sleeps.