MCP Server

Setup

You need a free Burn account (5 saves/day; Pro is $4.99/month for unlimited). Get your token in the app: Settings → MCP Server → Generate token → Copy.

Claude Desktop

{
  "mcpServers": {
    "burn": {
      "command": "npx",
      "args": ["burn-mcp-server"],
      "env": { "BURN_MCP_TOKEN": "<your-token>" }
    }
  }
}

Claude Code (CLI)

claude mcp add burn -e BURN_MCP_TOKEN=<your-token> -- npx burn-mcp-server

Cursor

Add the same JSON block to .cursor/mcp.json. Windsurf and any MCP-compatible client work identically.

What the 26 tools cover

Full tool reference lives in the GitHub README.

Example prompts

• “Go through my Flame inbox, read each article, keep anything about AI agents and burn the rest.”
• “Search my Vault for everything on context engineering and summarize the main schools of thought.”
• “Create a collection called ‘MCP in production’ from my saves, then write an overview of what my sources agree and disagree on.”
• “Add a watched source for Simon Willison’s blog so new posts land in my Flame.”
• “I’m writing about read-later apps. Pull every relevant bookmark I have and list the claims I can support with a source.”

Burn MCP vs Raindrop MCP vs Readwise MCP

Three ways to give an AI agent access to your saved reading, honestly compared (verified July 2026):

The honest split: Readwise MCP is the pick if your value lives in years of highlights — it’s hosted, official, and highlight-native. The community Raindrop MCP servers make the agent a librarian for a large bookmark archive. Burn MCP is the only one built around a triage loop: the agent doesn’t just search your pile, it works the 24-hour queue with you — reading, keeping, and burning — and can auto-feed that queue from watched sources. If the pile is the problem, comparison over; if the pile is the point, pick by where your data already lives. Deeper dives: Burn vs Readwise · Burn vs Raindrop.

FAQ