Using Fuzio documentation in AI coding tools

The Fuzio Java API reference is available on Context7 — a retrieval service that provides up-to-date documentation to AI coding tools. Once Context7 is connected to your AI coding tool, the AI assistant automatically looks up real API signatures and descriptions before generating Fuzio code.

Choose either way to connect:

• One-command setup;
• Manual MCP server configuration.

One-command setup 

If you use Cursor, Claude Code, Codex, Opencode, or Antigravity, we recommend the official Context7 installer:

npx ctx7 setup

The setup involves two steps:

1. How should your agent access Context7? — we recommend choosing MCP.
2. Sign in to Context7 — the terminal shows a one-time code and a link. Open the link in a browser, sign in, and approve. This is a one-time action: the installer then generates an API key and writes the configuration for you, so no further sign-in is required.

To uninstall, run npx ctx7 remove.

Manual configuration 

Context7 connects to AI coding tools via MCP. The remote endpoint is https://mcp.context7.com/mcp. Alternatively, you can run the server locally with npx -y @upstash/context7-mcp.

Cursor 

The fastest way is to click the one-click install link, and Cursor completes the configuration automatically. To configure manually:

1. Open Cursor.
2. Go to Settings → Cursor Settings → Tools & MCP.
3. Click New MCP Server and add the following configuration. The global configuration file is ~/.cursor/mcp.json:

{
  "mcpServers": {
    "context7": {
      "url": "https://mcp.context7.com/mcp"
    }
  }
}

4. Save the configuration and make sure the server is enabled.

Claude Code 

Open a terminal and run the following command:

claude mcp add --scope user --transport http context7 https://mcp.context7.com/mcp

Run claude mcp list to confirm that context7 appears in the list.

Trae 

Trae has a built-in MCP marketplace, so you can add Context7 directly from it:

1. Open the settings window: in the IDE mode, click the Settings icon in the top-right corner; in the SOLO mode, click the Settings icon in the top-right corner of the chat panel.
2. Select MCP in the left navigation bar to open the MCP window.
3. Click Add → Add from Marketplace in the top-right corner of the MCP window. If this is your first MCP server, you can also click the Add from Marketplace button in the middle of the window. This takes you to the MCP marketplace in TraeCode.
4. Find Context7 and click the + button next to it.
5. Click Confirm to finish.

JetBrains AI Assistant 

1. In your JetBrains IDE, go to Settings → Tools → AI Assistant → Model Context Protocol (MCP).
2. Click + Add and choose to import as JSON:

{
  "mcpServers": {
    "context7": {
      "url": "https://mcp.context7.com/mcp"
    }
  }
}

3. Click OK and Apply to save. Running the server locally requires Node.js 20 or later.

VS Code (GitHub Copilot) 

Install the Context7 extension from the marketplace, or add the server manually in your project’s .vscode/mcp.json:

{
  "servers": {
    "context7": {
      "type": "http",
      "url": "https://mcp.context7.com/mcp"
    }
  }
}

The complete configurations with an API key are shown below. Replace YOUR_API_KEY with your own key.

Remote (Cursor, Trae, JetBrains AI Assistant, and others):

{
  "mcpServers": {
    "context7": {
      "url": "https://mcp.context7.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Local (running the MCP server locally via npx):

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

This guide covers only the most common tools. If you use another client or need advanced configuration, see the official Context7 documentation: MCP Clients, which provides remote and local configuration examples for more than 30 clients.

Using it in a conversation 

Once configured, append use context7 to your prompt, and the AI assistant retrieves the documentation before answering:

Embed a browser into a Swing app with Fuzio and load a web page. Give me complete runnable code. use context7

Available tools 

Once configured, the AI assistant can call two MCP tools. Usually no manual intervention is required.

resolve-library-id 

Resolves a library name mentioned in natural language into a Context7 library ID, so simply mentioning Fuzio in your prompt is enough.

query-docs 

Retrieves documentation snippets of the given library by question and returns real API signatures, descriptions, and source links. Example prompts:

• How do I embed a browser into a Swing app with Fuzio and load a web page?
• How do I intercept network requests and modify request headers with Fuzio?
• How do I intercept file downloads in Fuzio and customize the save location?

Zero-configuration option 

Don’t want to change your IDE configuration, or your tool doesn’t support MCP? You can paste a Context7 documentation link directly into an AI conversation.

Context7 provides an llms.txt documentation endpoint for every library. The tokens parameter limits the length of the returned content:

https://context7.com/websites/fuzio_tech_api_2026_6_0_java/llms.txt?tokens=5000

Paste this link into any AI conversation with web access — Cursor, Claude, ChatGPT, and so on — and ask the AI to read the documentation before answering.

For example:

Read this Fuzio API documentation first, then write the code for me:
https://context7.com/websites/fuzio_tech_api_2026_6_0_java/llms.txt?tokens=5000

Using the official Fuzio llms.txt 

Besides Context7, the Fuzio website also provides a documentation index for AI tools that follows the llms.txt specification:

https://fuzio.tech/llms.txt

It indexes all documentation pages of the website, organized into quick start, guides, and release notes.