Desk3 MCP Server

Instructions

Implementation Reference

• src/desk3_service/server.py:210-219 (handler)

  The primary handler function for the 'get_cycles' tool. It fetches cryptocurrency market cycle indicators (Puell Multiple, Pi Cycle Top, likelihood) from the external API endpoint.

  async def get_cycles() -> dict[str, Any]:
      """
      Get Simple indicators data including Puell Multiple Status, Pi Cycle Top Status, and Crypto Market Cycle Top Indicator.
      :return: Simple indicators data with puellMultiple, piCycleTop, and likelihood fields
      """
      url = 'https://mcp.desk3.io/v1/market/cycles'
      try:
          return request_api('get', url)
      except Exception as e:
          raise RuntimeError(f"Failed to fetch cycles data: {e}")

• src/desk3_service/server.py:675-679 (schema)

  The JSON schema definition for the 'get_cycles' tool input, which requires no parameters.

  inputSchema={
      "type": "object",
      "properties": {},
      "required": [],
  },

• src/desk3_service/server.py:672-680 (registration)

  Registration of the 'get_cycles' tool in the MCP server's list_tools handler, including name, description, and schema.

  types.Tool(
      name="get_cycles",
      description="Does the Bitcoin Four-Year Cycle Exist? Discover the cryptocurrency market cycle indicator that helps you identify the top of the cryptocurrency bull market. This is a collection of publicly available signals including Pi Cycle and Puell Multiple data. Return fields: (puellMultiple Puell: multiple status / piCycleTop: Pi cycle top status / likelihood: cryptocurrency market cycle top indicator) ",
      inputSchema={
          "type": "object",
          "properties": {},
          "required": [],
      },
  ),

• src/desk3_service/server.py:872-882 (handler)

  The MCP tool call dispatcher case that invokes the get_cycles handler and formats the response as TextContent.

  case "get_cycles":
      try:
          data = await get_cycles()
          return [
              types.TextContent(
                  type="text",
                  text=json.dumps(data, indent=2),
              )
          ]
      except Exception as e:
          raise RuntimeError(f"Failed to fetch cycles data: {e}")