Quick Start

Getting Started: MCP + ClawdBot in 10 Minutes

Follow these steps to get a fully working ClawdBot with two essential MCPs: Filesystem (scoped) and Brave Search.

Step 1 — Install ClawdBot

curl -fsSL https://clawd.bot/install.sh | bash

Step 2 — Add MCP servers

Open your ClawdBot config and paste the template below. Replace paths and keys.

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "./workdir"]
    },
    "brave-search": {
      "url": "https://api.search.brave.com/mcp/",
      "headers": {"Authorization": "Bearer ${BRAVE_API_KEY}"}
    }
  }
}

• Set ./workdir to your project folder.
• Store BRAVE_API_KEY in environment variables, not plaintext.

Step 3 — Verify

• clawdbot status should show MCP servers running.
• Ask: “Read README.md and summarize” to verify Filesystem scope.
• Ask: “Search latest MCP news” to verify Brave Search.

Next

Move to the Top 10 list to add GitHub, PostgreSQL, Google Drive, and more.