mcp▌

Use structured MCP tools (get_scene_info, screenshot) for quick inspection.

Companion CLI scripts for MCP servers with file I/O, batch processing, and local caching. \n \n Provides one script per MCP tool with extended capabilities: batch input/output files, result caching, multiple output formats (JSON, CSV, table), and parallel processing \n Uses consistent argument patterns across all scripts ( --input , --output , --format , --profile , --verbose ) for predictable developer experience \n Scripts output structured JSON by default for easy parsing by Claude Code, with

Deploy remote MCP servers on Cloudflare Workers with TypeScript, OAuth, and Durable Objects. \n \n Supports SSE and HTTP Streamable transports with automatic WebSocket hibernation for stateful sessions; choose from five auth templates (no-auth, bearer, GitHub/Google OAuth, enterprise SSO) \n Base path configuration is critical—server and client URLs must match exactly (e.g., serveSSE(\"/sse\") requires client URL https://worker.dev/sse , not https://worker.dev ) \n Includes 24 documented error s

Build MCP servers with TypeScript on Cloudflare Workers, preventing 20 documented production issues. \n \n Supports tools, resources, prompts, tasks, and three authentication patterns (API keys, OAuth, Zero Trust) with built-in Cloudflare service integrations (D1, KV, R2, Vectorize) \n Requires fresh McpServer instance per HTTP request and StreamableHTTPServerTransport for production; SSE transport is deprecated \n Prevents critical issues including server instance reuse breaking concurrent sess

Skill by ara.so — Daily 2026 Skills collection.

Build MCP servers with Spring AI using declarative tools, prompt templates, and native Spring integration patterns. \n \n Exposes Spring components as AI-callable tools via @Tool annotation, with parameter documentation through @ToolParam for AI model understanding \n Supports three transport modes (stdio, HTTP, SSE) with built-in Spring Security integration for role-based access control and audit logging \n Includes reusable prompt templates using @PromptTemplate , dynamic tool registration, mu

User Input:

Build and deploy MCP servers in Python using FastMCP with tools, resources, and prompts. \n \n Scaffolds a working Python MCP server from a description; supports tools (callable functions), resources (readable data), and prompts (reusable templates) \n Includes local testing modes (direct run, dev mode with auto-reload, HTTP transport) and MCP Inspector integration \n Deploys to FastMCP Cloud, Docker, or Cloudflare Workers; pre-deploy checklist catches common issues like missing module-level ser

이 스킬이 호출되면 아래 STOP PROTOCOL을 반드시 따른다.

To create high-quality MCP (Model Context Protocol) servers that enable LLMs to effectively interact with external services, use this skill. An MCP server provides tools that allow LLMs to access external services and APIs. The quality of an MCP server is measured by how well it enables LLMs to accomplish real-world tasks using the tools provided.