Skip to main content

Connect to an Agent

The fastest path — the tea-rags-setup plugin registers the MCP server for you:

/plugin marketplace add artk0de/TeaRAGs-MCP
/plugin install tea-rags-setup@tea-rags
/tea-rags-setup:install

The wizard detects your environment, installs tea-rags, configures Qdrant and embeddings, and writes the MCP entry to your Claude Code config. No manual claude mcp add needed.

After the wizard finishes, install the Claude-only skills plugin:

/plugin install tea-rags@tea-rags

This is the final step — it gives your agent /tea-rags:explore, /tea-rags:bug-hunt, and the other skills. See Installation for the full step list.

Claude Code — Manual MCP Registration​

Use this only if you installed tea-rags manually or want to override the plugin's config.

# Local setup — Qdrant starts automatically (embedded), Ollama on localhost
claude mcp add tea-rags -s user -- tea-rags
tip

QDRANT_URL is autodetected: probes localhost:6333 for external Qdrant, falls back to embedded. No configuration needed for local setups.

claude mcp add flag order

-e KEY=VAL is a claude mcp add option, so env vars go before the -- separator. Everything after -- is the command and its args.

Remote Server Setup​

# External Qdrant + Ollama on separate host
claude mcp add tea-rags -s user \
  -e QDRANT_URL=http://192.168.1.100:6333 \
  -e EMBEDDING_BASE_URL=http://192.168.1.100:11434 \
  -- tea-rags

Qdrant Cloud​

claude mcp add tea-rags -s user \
  -e QDRANT_URL=https://your-cluster.qdrant.io:6333 \
  -e QDRANT_API_KEY=your-api-key-here \
  -- tea-rags

HTTP Transport (Remote)​

Security

When deploying HTTP transport in production:

  • Always run behind a reverse proxy (nginx, Caddy) with HTTPS
  • Implement authentication/authorization at the proxy level
  • Use firewalls to restrict access to trusted networks
  • Never expose directly to the public internet without protection

Start the server:

SERVER_TRANSPORT=http SERVER_HTTP_PORT=3000 tea-rags

Configure client:

{
  "mcpServers": {
    "qdrant": {
      "url": "http://your-server:3000/mcp"
    }
  }
}

Using a Different Provider​

"env": {
  "EMBEDDING_PROVIDER": "openai",
  "OPENAI_API_KEY": "sk-..."
}

See Configuration for all options.