For AI agents: the Yardstick MCP server

Yardstick Outfitters runs a free, public, read-only Model Context Protocol (MCP) server. Any AI assistant that supports remote MCP servers can ask it which size to buy, read our fit guides, or check verified ADV and dual-sport gear deals.

Server URL

https://yardstickoutfitters.com/mcp

• Transport: Streamable HTTP (JSON-RPC 2.0 over POST, JSON responses)
• Protocol versions: 2026-07-28 (current), plus 2025-11-25, 2025-06-18 and 2025-03-26 via initialize
• Auth: none. No personal data is stored, and nothing can be bought through it.
• Limits: about 60 requests per minute per IP.

Tools

• size_recommendation: give a brand, item (jacket, pants, boots, helmet or gloves) and your measurements in inches or cm. You get a recommended size and fit notes from our published guides. Covered today: Alpinestars (jackets, pants), Dainese (jackets), Klim (jackets), REV'IT! (jackets, pants), Shoei (helmets), Gaerne (boots). For anything else it says plainly that it isn't covered. It never guesses.
• list_fit_guides: all our fit guides, with links and one-line summaries.
• current_deals: verified ADV and dual-sport deals with price, regular price, retailer link, an "as of" time, and optional made_in_usa (set only from official brand sources under the FTC all-or-virtually-all standard). Filter by category, brand, max price or made_in_usa.
• about_yardstick: what Yardstick does, plus the waitlist and policy links.

Every answer includes a link back to the relevant page on this site and a short attribution line.

Add it to your assistant

Most MCP clients take a config like this:

{
  "mcpServers": {
    "yardstick": {
      "type": "http",
      "url": "https://yardstickoutfitters.com/mcp"
    }
  }
}

In apps with a "custom connector" or "add MCP server" screen, paste the URL above.

Example request

curl -s https://yardstickoutfitters.com/mcp \
  -H 'Content-Type: application/json' -H 'Accept: application/json, text/event-stream' \
  -H 'MCP-Protocol-Version: 2026-07-28' -H 'Mcp-Method: tools/call' -H 'Mcp-Name: size_recommendation' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"size_recommendation",
  "arguments":{"brand":"Klim","product_type":"jacket","units":"in","measurements":{"chest":42}},
  "_meta":{"io.modelcontextprotocol/protocolVersion":"2026-07-28"}}}'

Size suggestions are a best estimate from brand charts and rider reports, not a guarantee. See our Terms. Plain-text summaries for AI tools are in llms.txt.