[Keywords: MCP, MCPmarket, Claude, Cursor, AI Agent, Cloud Hosting MCP, Model Context Protocol, MCP Server deployment, Claude plug-in market, MCP Playground, agent development, LLM tool call, AI tool market, A2A protocol]
In the wave of agents gradually becoming mainstream, a platform called "is quietly changing the way developers interact with AI tools.
Recently, it announced the completion of a major upgrade: the official launch of cloud hosting and deployment services, and the native Playground experimental field will be launched soon, realizing a seamless experience of "one-click deployment + dialogue as test". This upgrade marks the official entry of the MCP Server ecosystem into a new era of "zero code and zero environment configuration".
What is MCP?
MCP, full name Model Context Protocol, is a model context calling protocol driven by leading companies such as Anthropic and OpenAI. It aims to enable large language models (LLMs) to call various external tools and data sources safely and structured.
If Function Calling is a remote assistant built for a single task, then MCP is more like a set of plug-in ecological standards built for LLM, with greater flexibility and versatility.
One-click deployment: Cloud Hosted MCP Server is here!
In the past, developers often needed to deploy an MCP Server:
• Local pull code
• Installation environment dependencies
• Setting up API keys
• Debug local ports
• Testing with client connections such as Claude/Cursor/OpenDevin
This process is almost a "dissuasion" level for users with non-engineering backgrounds. But now, MCPmarket's cloud hosting service makes it all exceptionally simple:
Just click on "Generate Link" and you can call LLM platforms such as Claude and Cursor in one click to perform ready-to-use and test. No need to write a line of code.
The first batch of popular high-frequency services have been launched:
• MiniMax multimodal generation: supports text, pictures, and voice multimodal input and output
• Gaode Maps & Google Maps: Address Positioning, Route Planning, Distance Calculation
• Playwright & Firecrawl: Automatic web crawling, simulate browser operations
• Perplexity & Brave Search: Real-time search + Q&A
• Sequential Thinking: Multi-step logical reasoning
• GitHub MCP: Obtain PR, Issue, Project Data
• Time: Time conversion and time zone processing
• MarkItDown / Fetch: Markdown conversion and web structure analysis
These services have been opened for use in the "Cloud Hosting Selection" column on the homepage, covering multiple key Agent scenarios such as maps, searches, generations, crawls, and conversions, covering almost most of the underlying capability calls required by LLM Agents.
The "super plug-in market" of MCP tools is taking shape.
Playground Experimental Field: Dialogue is testing, truly "zero threshold"
MCPmarket is also about to release a new module: Playground Laboratory.
This is an AI conversation environment that supports MCP natively. Users can debug any MCP services without any deployment. They just need to chat with the AI on the web page.
For example, you only need to enter:
🗨️ "What is the distance from Shanghai to Beijing?"
Playground will make LLM automatically call [Gaod Map MCP] to return accurate mileage data.
: Not only the market, but also a new paradigm for the development of intelligent bodies
MCPmarket is not only a tool market for MCP Server, but is also becoming a one-stop intelligent development infrastructure:
• Supports MCP Server cloud hosting (Serverless)
• Provide a Playground conversational test site
• Aggregate tens of thousands of MCP tool components
• We are building docking capabilities with the A2A protocol ecosystem to realize Agent interconnection
MCPmarket is opening the last mile from "Model Call Tool" to "Cooperation between Agents".
Official website:/
Official account: @MCP Planet
If you are a developer, an AI product manager who is interested in AI Agent, or a novice who is studying LLM tool calls, this upgrade is worthy of your attention at the first time.