发现优秀的 MCP 服务器

hwatu

Verification browser for coding agents. Headless-by-default WebKit windows with DOM eval, screenshots, pixel-diff with a real match percentage and heatmap, and live hand-off to a human. One static binary, no Chromium.

winuse-mcp

An MCP server that gives Claude Code on Windows the ability to see the screen and control the mouse and keyboard, mirroring Anthropic's computer use tool for screenshot, click, type, and scroll actions.

debugbase-mcp

Enables AI agents to search and submit solutions, ask and answer questions, share findings, and vote on content via DebugBase's collective knowledge base.

research-mcp

A stateless MCP server that provides web search, single page reading, and batch page reading tools, with automatic failover across multiple search and content extraction providers.

Israel Grocery MCP

Enables cross-store price comparison and recipe-driven cart automation for Israeli grocery stores Shufersal and Tiv Taam, with an extensible architecture for additional stores.

Larry

Social coding platform for AI agents. Post snippets, fork code, vote, follow agents, build reputation. REST API + MCP server.

Firecrawl MCP Server

一个模型上下文协议服务器，它使 AI 助手能够通过 Firecrawl API 执行高级网络抓取、爬行、搜索和数据提取。

MCP Thinking Server

A powerful MCP server that enhances LLMs with advanced sequential thinking capabilities, supporting 19 thinking modes for structured reasoning and complex cognitive tasks.

daedalus

Enables driving a real Chrome instance remotely through MCP, including per-tab control, JavaScript evaluation, screenshots, cookie and network capture, and CDP fallback with token-based authentication.

Aseprite MCP Tools

A Python MCP server enabling programmatic interaction with Aseprite for pixel art creation and manipulation with features like drawing operations, palette management, and batch processing.

MCP Workspace Server

Provides secure, sandboxed file system access for AI assistants to read, write, and manage project files with controlled command execution capabilities, all confined to a designated workspace directory.

ae-mcp

Enables AI assistants to operate Adobe After Effects with the dexterity of an editor: reading projects, creating and rigging layers, setting real velocity curves, animating text, building native SVG-based shapes, applying effects, expressions, masks, and viewing render snapshots to verify results.

influx-mcp

MCP server for InfluxDB 2.x, tuned for Home Assistant long-term storage, enabling reading sensor history, finding anomalies, and running Flux queries.

text2img-mcp

Creates AI-generated images from text prompts using Google Gemini API, supporting business use cases like proposals, infographics, and charts.

EasyPlaywrightMCP

Enables automated testing and demo video creation using Playwright, with login, session orchestration, narration, and video compilation.

onbid-mcp

Enables LLM clients to ask plain-language questions about Korean public auction property data from the 온비드 OpenAPI, including normalized pricing, location, and failed-sale history.

Zignet

Enables AI-powered Zig programming assistance through code generation, debugging, and documentation explanation. Uses local LLM models to provide idiomatic Zig code creation and analysis capabilities.

Frappe Assistant Core

MCP server that enables LLMs to interact with ERPNext/Frappe sites for document CRUD, search, reports, workflows, and analytics, respecting user permissions and logging all actions.

oura-mcp

An MCP server that enables AI assistants to access Oura Ring health data including activity, sleep, heart rate, stress, and personal info.

Mercantry MCP Server

Enables AI agents to search and retrieve structured merchant data, check availability, manage bookings, and submit feedback against an open commerce registry with real restaurant data for LA, Hong Kong, and Tokyo.

instantKOM MCP Server

Enables AI assistants to manage instantKOM messenger platform resources (channels, contacts, messages, newsletters, bots, flows, analytics) via structured type-safe tool calls.

ShortsMonkey MCP

A read-only MCP server for discovering YouTube outliers, analyzing viral Shorts, and evaluating single-video performance using stored snapshots.

market-recap

Provides a one-call snapshot of recent market activity and events.

mcp-better

A modern MCP server demonstrating the 7/28 protocol era, offering health and echo tools with stamped list caching and stable ordering over stdio or Streamable HTTP.

tokencost-mcp-server

An MCP (Model Context Protocol) server that provides real-time LLM token pricing data for 60+ AI models across 15 providers.

Generate-Prd-Prompt

Mercury Spec Ops MCP Server is a dynamic prompt generation and template assembly tool based on a modular architecture. It is suitable for the interaction between AI assistants and professional content, and supports the dynamic generation of 31 technology stacks, 10 analysis dimensions and 34 templat

dida-mcp-server

Enables AI assistants to manage TickTick/Dida365 tasks, projects, and tags through the MCP protocol, with features for GTD-based task organization and OAuth authentication.

Controtto

Okay, I understand. I will do my best to translate English text into Chinese, and when presented with Go code, I will attempt to analyze it from the perspective of Domain-Driven Design (DDD) and Clean Architecture principles. **Specifically, when analyzing Go code, I will look for:** * **DDD Aspects:** * **Ubiquitous Language:** Is the code using terminology that aligns with the business domain? Are the names of variables, functions, and types meaningful to domain experts? * **Entities:** Are there well-defined entities with identity and behavior? * **Value Objects:** Are there immutable value objects representing domain concepts? * **Aggregates:** Are aggregates used to enforce consistency and manage transactions? Are aggregate roots clearly defined? * **Domain Services:** Are domain services used to encapsulate complex domain logic that doesn't naturally belong to an entity or value object? * **Repositories:** Are repositories used to abstract data access and persistence? * **Domain Events:** Are domain events used to decouple different parts of the system and react to changes in the domain? * **Bounded Contexts:** (If applicable) Is the code organized into bounded contexts with clear boundaries and responsibilities? * **Clean Architecture Aspects:** * **Dependency Inversion Principle (DIP):** Are high-level modules not dependent on low-level modules? Are abstractions used to decouple layers? * **Interface Segregation Principle (ISP):** Are interfaces small and focused, avoiding unnecessary dependencies? * **Single Responsibility Principle (SRP):** Do classes/modules have a single, well-defined responsibility? * **Layers:** Are there distinct layers (e.g., presentation, application, domain, infrastructure)? * **Use Cases/Interactors:** Are use cases clearly defined and implemented as interactors? * **Entities (Domain Layer):** Is the core business logic (entities and domain rules) independent of frameworks and infrastructure? * **Frameworks & Drivers (Outer Layer):** Are frameworks and infrastructure concerns kept separate from the core business logic? * **Testability:** Is the code easily testable, with clear separation of concerns? **My analysis will be strict and critical.** I will point out potential violations of these principles and suggest improvements. I will also consider the trade-offs involved in applying these principles, as strict adherence is not always practical or beneficial. **Important Considerations:** * **Context is Key:** My analysis will be limited by the information you provide. The more context you give me about the domain, the requirements, and the overall architecture, the better I can assess the code. * **Subjectivity:** DDD and Clean Architecture are not rigid rules, but rather guidelines. There is room for interpretation and different approaches. My analysis will reflect my understanding of these principles, which may differ from yours. * **Code Completeness:** I can only analyze the code you provide. I cannot infer the entire system architecture or domain model from a small snippet. **Example:** If you provide me with the following Go code: ```go package main import "fmt" type User struct { ID int Name string Email string } func SaveUser(db *sql.DB, user User) error { _, err := db.Exec("INSERT INTO users (name, email) VALUES (?, ?)", user.Name, user.Email) return err } func main() { // ... database connection ... user := User{Name: "John Doe", Email: "john.doe@example.com"} err := SaveUser(db, user) if err != nil { fmt.Println("Error saving user:", err) } } ``` I might respond with something like: "This code has several potential issues from a DDD and Clean Architecture perspective: * **Lack of Domain Layer:** The `User` struct is directly coupled to the database. It should be part of a dedicated domain layer, separate from infrastructure concerns. * **Anemic Domain Model:** The `User` struct only contains data and lacks behavior. Consider adding methods to the `User` struct that encapsulate domain logic related to users. * **Direct Database Access:** The `SaveUser` function directly interacts with the database. This violates the Dependency Inversion Principle. A repository interface should be used to abstract data access. * **Missing Repository:** There is no repository interface or implementation to handle user persistence. * **No Use Case:** The `main` function directly calls the `SaveUser` function. This should be encapsulated in a use case or interactor. To improve this code, consider: 1. Creating a `domain` package with the `User` entity and any related value objects. 2. Defining a `UserRepository` interface in the `domain` package. 3. Implementing the `UserRepository` interface in an `infrastructure` package, using a database connection. 4. Creating an `application` package with a `CreateUser` use case that uses the `UserRepository` to persist the user. This would result in a more decoupled, testable, and maintainable codebase that adheres to DDD and Clean Architecture principles." **Now, please provide me with the text you want me to translate or the Go code you want me to analyze.**

spd-mcp

Enables remote control of Siglent SPD3303X DC power supplies via SCPI over TCP/IP, supporting output control, measurement, timers, and network configuration.