Give your AI agent a naming tool that verifies.

Any model can suggest a name. None can reliably tell you the .com is free or the trademark is clear. This MCP server can. Connect any agent and it generates brandable names whose .com is available right now, checks domains authoritatively, and screens the one you pick against the US and EU trademark registers. Free during the beta.

The tools

• generate_names brandable names whose .com is available right now, each checked live against domain records. Free.
• check_domain authoritative .com availability via the Verisign RDAP registry. Free.
• screen_trademark a US (USPTO) and EU (EUIPO) registry knockout search with exact, phonetic, and fuzzy matching weighted by Nice class. Returns a green, yellow, red, or unknown band with the top matches. Metered.
• score_name a 0 to 100 defensibility score combining the .com check and the trademark screen, with a confidence flag. Metered.

Connect

Point any Streamable HTTP MCP client (Claude, Cursor, your own agent) at the endpoint and pass your key as a bearer token.

{
  "mcpServers": {
    "startupnamegenerator": {
      "url": "https://startupnamegenerator.com/api/mcp",
      "headers": { "Authorization": "Bearer sng_your_key" }
    }
  }
}

Endpoint: https://startupnamegenerator.com/api/mcp. Every request needs a valid key. There is also a local stdio build for running the tools with your own provider keys.

Get a free key

One email, one key, no card. Free-tier limits: 30 generations and 100 domain checks per day, 25 trademark screens per month.

Pricing

Free during the beta. After beta, generation and domain checks stay free; trademark screening moves to usage-based credit packs. Whatever you build on the free tier keeps working.

screen_trademark and score_name run an automated knockout search of the USPTO and EUIPO registries. They are not legal advice and not a clearance opinion, and never report a name as cleared or safe. Registries change daily and results are dated. Before filing, have counsel run full clearance.