Claude Desktop インストールガイド
ローカル MCP 設定に対応するデスクトップクライアント。
MCP Servers
Apex Log MCP Server - AI-powered Salesforce Apex debug log analysis. Find performance bottlenecks, slow methods, SOQL bottlenecks, and governor limit issues.
- 小爱音箱设备控制（支持40+款型号） - 涂鸦设备精准控制 - 场景联动执行
Use DeepSeek from Claude Code, Codex, or any MCP-compatible client as a small, cheap supervised worker.
Illustrative estimates. The ~90% is Claude-side token reduction on analysis-heavy tasks — not total cost: Kimi's own subscription still applies.
Observal is self-hosted registry for your coding agent extensions with a built in insight engine. Setup Observal, define the scope and share your Skills, MCPs and Agents with your peers.
mcp-difyworkflow-server is an mcp server Tools application that implements the query and invocation of Dify workflows, supporting the on-demand operation of multiple custom Dify workflows.
Reqable MCP Server is a Dart-based MCP server that communicates with AI tools over stdio and exposes Reqable capabilities through MCP tools.
Production-grade vector memory MCP server for Claude Code. Semantic search, per-project isolation, rules enforcement, team sharing via git.
A monorepo of Unraid tooling: two MCP servers (Python and Rust) and three Unraid OS plugins, plus the Claude/Codex agent integrations that surface them.
Dinos MCP Server, make your code, on MCP Action and execute by AI
Garmin Connect MCP server based on garth.
CROW - Your AI Agent. MCPs, OpenRouter, Any Model or local. It's your choice.
Agent Skills
Molecode presents molecules as code and enables LLMs to operate and reason on chemistry directly. Instead of making language models reconstruct molecular structure from cryptic strings, MoleCode lets...
An agent skill for managing isolated execution environments
The skills I actually use to build software with coding agents. Straight out of my .claude/skills/ folder.
The #1 Kubernetes skill for Claude Code and Codex, measured by GitHub stars.
ProductSpec is the open standard for product intent. It acts as a Product Harness for AI-native software work: what to build, what not to build, how to prove completion, and when intent changes.
AI agents like Claude Code ship features fast. AIDD Framework keeps them working, secure, and maintainable.
中文内部技术写作的 agent skill。约束设计文档、评审稿、postmortem、分享稿等场景的语气、句法、结构，让产出贴近技术汇报，避开营销稿和散文气质。
Local-first harness for agent-built codebases. Humans steer, agents execute, maestro is the substrate.
一句话：输入一个基金代码，Agent 把基金经理 8 年的每一笔买卖翻出来，拿 12 个月后的真实走势逐笔验对错，最后吐出一份 700KB 的单文件交互报告——K 线战役回放、买卖点验尸、独立性审判、造神检测，全都有。
An AI Agent skill that wraps the Bitget Wallet API, enabling natural-language-driven on-chain data queries and swap operations.
A multi-agent marathon simulation built with Google ADK and Gemini. AI agents plan a marathon route through Las Vegas, simulate the environment around it (weather, traffic, crowds), and run the race...
Together they produce a structured Markdown test plan and a findings report with 10-state verdicts and an explicit SUT / harness / checker / environment blame classification.