reddit-mcp

A Model Context Protocol (MCP) server that lets LLMs fetch and read content from Reddit. Give Claude the ability to browse posts, comments, and subreddit listings.

Features

• Post pages — fetches title, body, metadata, and threaded comments

• Subreddit listings — retrieves post listings from any subreddit

• User pages — browses a user's post history

• Short links — automatically expands redd.it URLs

• Safety built-in — untrusted content disclaimer, text truncation, timeout protection, rate limit handling

Requirements

• Node.js 22+

Installation

git clone https://github.com/quzma/reddit-mcp.git
cd reddit-mcp
npm install
npm run build

Usage

Claude Desktop

Add to your Claude Desktop MCP config (claude_desktop_config.json):

{
  "mcpServers": {
    "reddit": {
      "command": "node",
      "args": ["/absolute/path/to/reddit-mcp/dist/index.js"]
    }
  }
}

Claude Code

Add to your Claude Code MCP settings:

claude mcp add reddit -- node /absolute/path/to/reddit-mcp/dist/index.js

Development

npm run dev

Tool

fetch_reddit

Fetch content from a Reddit URL.

Supported URL formats:

• https://reddit.com/r/subreddit/comments/id/title/

• https://reddit.com/r/subreddit/

• https://reddit.com/user/username/

• https://redd.it/id

• Old/new/www Reddit URL variants

How It Works

Uses Reddit's public JSON API — no API keys or OAuth needed. The server communicates over stdio using the MCP protocol, so it works as a local tool for any MCP-compatible client.

Content safety measures:

• All output is prefixed with an untrusted-content disclaimer

• Post/comment text truncated at 20,000 characters

• Comment nesting limited to 2 levels, 5 replies per comment

• 15-second request timeout

• Rate limit (429) and forbidden (403) error handling

License

MIT — see LICENSE