# tendril-mcp

A [Model Context Protocol](https://modelcontextprotocol.io) server that gives an
AI agent three tools backed by the Tendril API (`https://www.ten-dril.com`):

- `tendril_scrape` — fetch a page → clean Markdown + structured data + links
- `tendril_map` — discover a site's URLs (sitemaps + robots + homepage + llms.txt)
- `tendril_status` — check which fetch tiers are up

It's a single self-contained file — **no `npm install` needed**, only Node ≥ 20.

## Install

Save `tendril-mcp.mjs` somewhere, then point your MCP client at it.

**Claude Desktop** — `~/Library/Application Support/Claude/claude_desktop_config.json`
(macOS) or `%APPDATA%\Claude\claude_desktop_config.json` (Windows):

```json
{
  "mcpServers": {
    "tendril": {
      "command": "node",
      "args": ["/absolute/path/to/tendril-mcp.mjs"],
      "env": { "TENDRIL_BASE_URL": "https://www.ten-dril.com" }
    }
  }
}
```

**Claude Code** — from your project or user scope:

```bash
claude mcp add tendril -- node /absolute/path/to/tendril-mcp.mjs
```

Restart the client; you should see `tendril_scrape`, `tendril_map`, `tendril_status`.

## Configuration

- `TENDRIL_BASE_URL` — API base (default `https://www.ten-dril.com`; set to a local
  deployment like `http://127.0.0.1:8092` for testing).
- `TENDRIL_API_KEY` — optional Bearer token (not required today).

Author: Simon-Pierre Boucher — contact@spboucher.ai