As described in the article, where do you configure MCP servers for shared team use in a Claude Code project?

In .mcp.json in the project root (committed to git for team sharing)

In the CLAUDE.md file at the project root

In .mcp.json in the project root (committed to git for team sharing)

In the system environment variables

🔌 MCP lets Claude Code talk to Jira, Slack, Google Drive, and anything with an API

The Model Context Protocol (MCP) is an open standard that connects Claude Code to external data sources. Instead of copy-pasting context from other tools, you give Claude direct access.

With MCP servers, Claude can: read your design docs from Google Drive, update tickets in Jira, pull messages from Slack, query your database, or call your own custom APIs. Each MCP server exposes specific "tools" that Claude can invoke during a session.

Configuration is simple — add servers to .mcp.json in your project root (shared with team) or ~/.claude.json (personal). Each server defines its command, arguments, and which tools it exposes.

The beauty: MCP servers are composable. You can connect multiple servers simultaneously. Claude sees all available tools and uses them as needed. If you want Claude to "check the Jira ticket, read the related design doc, and implement the feature" — MCP makes that a single prompt.