Claude Desktop 安装指南
支持本地 MCP 配置的桌面客户端。
MCP Servers
Self-hosted audit trail for AI agents — one command or one dashboard click from any step back to root cause.
Transparent MCP proxy that reduces the token cost of large upstream tool responses for nearly every other MCP server imaginable.
Power your chat with B2B data to create lead lists, research companies, personalize your outreach, and more.
The enterprise-grade TypeScript framework for building production-ready MCP servers. Decorators. Dependency Injection. Widgets. One framework to ship AI-native backends.
Streamable HTTP MCP server for Claude.ai custom connectors.
The purpose of the repo is to teach you how to make MCP servers. The target audience are non-technical with no real coding experience.
- — MCP Protocol v2 (spec revision 2026-07-28, streamable HTTP) and experimental StrongDM API support are here — see the release notes.
ChatSpatial replaces ad-hoc LLM code generation with . Instead of generating arbitrary scripts, the LLM selects tools and parameters from a curated registry, making spatial transcriptomics workflows...
Fictional concept interface—not an actual Codex UI or a measured context, token, speed, or latency benchmark.
Local-first creative studio with 37 model routes, transparent costs, local archives, and MCP access.
directory for Awesome MCP Servers
🤝 Show your support - give a ⭐️ if you liked the content
Agent Skills
Korean e-commerce 상세페이지(상폐) generator — 13 sections + 1080×12720 combined image via Codex OAuth. No OpenAI API key required.
梧桐雨的个人站点：Astro 5 静态生成，黑白主题｜My Personal Astro Blog
Shared skills for Agents
Autofish gives any agent a stable, explicit, and observable Android execution layer: inspect the screen, act through stable refs instead of brittle coordinates, verify outcomes after every step, and...
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。
A Claude Code skill that orchestrates a live agent team for strategic decisions. A team lead runs intake, classifies the question, spawns 3-4 teammates that work McKinsey-style frameworks in...
An atelier for your Agent: spec-driven workflows, deep thinking, and code quality.
Claude Code plugins for integrating with AI coding assistants.
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.