MCP server (AI assistants)

The FalconryScale MCP server lets an AI assistant work with your falconry records. Ask "How has Sable's weight moved this week?" or "Log 742 g for Sable, empty, this morning" and the assistant reads or writes your data directly, with the same permissions you have in the app.

How it worksCopy Anchor Link

MCP (Model Context Protocol) is an open standard that lets AI assistants call tools on other services. FalconryScale publishes one MCP server:

https://app.falconryscale.com/mcp

1. You add the URL to your assistant once. Nothing else to paste.
2. The assistant asks you to sign in to FalconryScale and shows what it wants to do: read your data, or read and log entries. You approve or decline. See OAuth for the details.
3. The assistant sees a list of tools, such as list_raptors or log_weight, with a description of each.
4. When your question needs data, the assistant calls a tool. FalconryScale checks its access, runs the request as you, and returns the result.
5. Your rules still apply. Team roles, plan limits and per-bird permissions are the same as in the app. A teammate's bird you can only view stays read-only.

It speaks Streamable HTTP, the current MCP transport. Nothing to install on your computer: the server runs on FalconryScale.

Sign-in (OAuth) is the main way to connect. If your tool can't sign in, use an API key instead, see Connect with an API key below.

ToolsCopy Anchor Link

Connect with sign-in (OAuth)Copy Anchor Link

Connect with an API keyCopy Anchor Link

Use a key when your tool has no sign-in support, for scripts, or when you want a key you can rotate yourself.

1. In the web app, open Settings › Developer, choose Create key, give it a name (for example "Claude Desktop"), pick Read only or Read and write, and copy the key. It is shown once. See API keys.
2. Add it to your client as a bearer header:

{
  "mcpServers": {
    "falconryscale": {
      "command": "npx",
      "args": [
        "-y", "mcp-remote", "https://app.falconryscale.com/mcp",
        "--header", "Authorization: Bearer ${FALCONRYSCALE_API_KEY}"
      ],
      "env": { "FALCONRYSCALE_API_KEY": "paste-your-key-here" }
    }
  }
}

See and disconnect connected assistantsCopy Anchor Link

Settings › Developer › Connected apps lists every assistant that has used your account through sign-in, with its access level and when it was first and last used. An assistant appears there after its first request, not at the moment you approve it.

Choose Disconnect to stop FalconryScale accepting that assistant for your account straight away. Also remove the FalconryScale connector inside the assistant, so it stops trying. Allow again undoes a disconnect. API keys are listed and revoked separately, under API keys on the same page.

Keep it safeCopy Anchor Link

• Read the approval screen: it names the assistant asking for access. Decline anything you didn't start.
• Prefer read-only access unless you want the assistant to log entries.
• One connection or key per assistant, so you can cut one off without breaking the others.
• Treat an API key like a password. Don't paste it into shared chats, screenshots or public repositories.
• Assistants can make mistakes: check entries they log, and review a tool call before approving it.

TroubleshootingCopy Anchor Link

Limits and error formats are on Errors and limits. For the same operations over plain HTTP, see the REST API.