LABSCO SUMMARY

Retrieval is bounded by a crawl you can enumerate before you trust it.

The pair is a scoping call followed by a retrieval call — list-topics says what has been indexed, search returns ranked documents for a question within those topics. Coverage is whatever the crawler has taken in, and list-topics is what shows where that boundary sits.

What it is

A retrieval server over crawled web content: a topic listing, and a ranked search across the pages crawled under those topics

What you get

list-topics returns the subject areas that have crawled content behind them, so a question can be scoped before it is asked.
search takes a question and a list of topics and returns the top-K most relevant crawled documents.

Requirements

No credentials to supply; the crawled index is what the two tools query.

Setup effort

One command — npx -y @driflyte/mcp-server

See installation options

USE CASES

01

Grounding an answer in a subject area outside the model's training

Read which topics have crawled content behind them, then ask the question against those topics and take the ranked documents back.

list-topicssearch

02

Checking a topic is covered before you rely on it

The topic list is the coverage boundary — anything outside it has no crawled pages behind it.

list-topics

2 tools

Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.

`list-topics`

Returns a list of topics for which resources (web pages, etc ...) have been crawled and content is available. This allows AI assistants to discover the most relevant and up-to-date subject areas currently indexed by the crawler.

BEFORE YOU INSTALL

What to know first

Tools you get

2 tools when we connected. Both tools read; indexing happens on Driflyte's side. The API allows 100 requests per 5 minutes per IP, and search's topK is capped at 30 (default 10).

Requires

npx on PATH
Node.js 18+

CONTEXT COST

Light Standing cost

Rated by the size of the 2 tool definitions that ride along on every turn, not by how many there are

INSTALLATION

Connect your client

Server setup

npx

TERMINAL

npx -y @driflyte/mcp-server

Hosted endpoint

ENDPOINT URL

https://mcp.driflyte.com/mcp

Add this hosted endpoint to your MCP client.

MCP Clients

ChatGPT

Name: Driflyte

MCP Server URL: https://mcp.driflyte.com/openai

Authentication: No authentication

Trust Setting: Check I trust this application

Claude Code

CLAUDE CODE · TERMINAL

claude mcp add driflyte -- npx -y @driflyte/mcp-server

Claude Desktop

CLAUDE DESKTOP · JSON

{
  "mcpServers": {
    "driflyte": {
      "command": "npx",
      "args": [
        "-y",
        "@driflyte/mcp-server"
      ]
    }
  }
}

claude_desktop_config.json

GitHub Copilot coding agent

GITHUB COPILOT CODING AGENT · JSON

{
  "mcpServers": {
    "driflyte": {
      "type": "local",
      "command": "npx",
      "args": [
        "-y",
        "@driflyte/mcp-server"
      ]
    }
  }
}

Repository > Settings > Copilot > Coding agent > MCP configuration

Cursor

CURSOR · JSON

{
  "mcpServers": {
    "driflyte": {
      "command": "npx",
      "args": [
        "-y",
        "@driflyte/mcp-server"
      ]
    }
  }
}

~/.cursor/mcp.json (or .cursor/mcp.json in your project folder)

Gemini CLI

GEMINI CLI · JSON

{
  "mcpServers": {
    "driflyte": {
      "command": "npx",
      "args": [
        "-y",
        "@driflyte/mcp-server"
      ]
    }
  }
}

~/.gemini/settings.json

Smithery

SMITHERY · TERMINAL

npx -y @smithery/cli install @serkan-ozal/driflyte-mcp-server --client <YOUR_SMITHERY_CLIENT_NAME> --key <YOUR_SMITHERY_API_KEY>

VS Code

VS CODE · JSON

{
  "servers": {
    "driflyte": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@driflyte/mcp-server"
      ]
    }
  }
}

.vscode/mcp.json

Windsurf

WINDSURF · JSON

{
  "mcpServers": {
    "driflyte": {
      "command": "npx",
      "args": [
        "-y",
        "@driflyte/mcp-server"
      ]
    }
  }
}

~/.codeium/windsurf/mcp_config.json

Driflyte