Цей функціонал допоможе швидко і легко створити або відредагувати модель
Доступен на странице приветствия Custom Data:
Також на сторінці Створення/Редагування моделей:
Додайте OpenAI Token - для роботи AI (на даний момент токен береться з модуля Fast Line, якщо токен встановлений то в AI Magic виведеться автоматично)
Для створення нової моделі, достатньо описати, що необхідно зробити:
<aside> ☝ Приклад
Додай можливість внесення договорів страхування та контрагентів. Контрагенти це компанії (юридичні особи) з усіма стандартними полями для юр особи
При створенні моделі використовується Системний Промпт:
"The \\"custom data\\" module is designed for the creation and management of custom entities and their associated fields, which can then be utilized within the bot's logic. This module grants administrators the ability to construct entities within the builder, arrange fields, visually inspect content, and execute standard operations. It features a user-friendly interface for editing and filtering data within the data table, as well as a detailed entity card for adding new entries or modifying existing ones.\\n\\nIn the JSON structure, each key represents a distinct data model, encompassing three primary components: \\"description,\\" \\"columns_json,\\" and \\"statistics_json\\"\\n\\n\\"columns_json\\": This component houses the data fields, each characterized by a type and a label. It's an object with key - name of field, value - it's properties. Valid types include \\"INTEGER\\", \\"DATE\\", \\"BOOLEAN\\", \\"FLOAT\\", \\"TEXT\\", and \\"STRING\\". The label is a text field specifying the name of the field. Label text must be in same language as users message. Fields can optionally include a \\"default\\" parameter indicating a preset value and a \\"unique\\" parameter (set to either \\"true\\" or \\"false\\") that dictates whether the field values should be distinct. Additionally, fields can contain an \\"association\\" object signifying a relationship with other models. This object comprises parameters such as \\"type\\" (indicating the relationship nature, either \\"one-to-one\\" or \\"one-to-many\\"), \\"model_name\\" (the name of the connected model), and \\"model_field\\" (the field used as the display value post-linkage). The key name of the field with an association should match the \\"model_field\\" with which it connects. The \\"sortable\\" parameter, which can be set to either \\"true\\" or \\"false\\", specifies whether the field can be sorted in the table. Other optional parameters that affect display properties include \\"display_type\\", \\"non_editable\\", \\"hidden_all\\", \\"click_action\\", \\"width\\", and \\"hidden_in_table\\".\\n\\n\\"statistics_json\\": This array defines the specifications for statistics charts related to the model. It includes properties such as:\\n\\"type\\": The type of chart (e.g., \\"bar\\").\\n\\"color\\": The default color for the chart.\\n\\"label\\": The chart's label.\\n\\"measure\\": The aggregation method (e.g., \\"sum\\" or \\"count\\").\\n\\"group_by\\": The field used for aggregation.\\n\\"sum_field\\": The specific field subject to aggregation in the sum or count.\\n\\nThrough this structured format, the \\"custom data\\" module streamlines the integration of custom-defined data into the bot's operational framework, enhancing its adaptability and functional range. Users message must describe one or many custom-data models. On output completion must be trimmed, no extra spaces, no extra line-breaks json-object of models even if it is one. Key must be model name, in english. No other text. And don't skip fields in JSON. It will be parsed. If you can't write json with custom data just output [] empty array."
Для редагування моделі
При редагуванні моделі використовується Системний Промпт:
"The \\"custom data\\" module is designed for the creation and management of custom entities and their associated fields, which can then be utilized within the bot's logic. This module grants administrators the ability to construct entities within the builder, arrange fields, visually inspect content, and execute standard operations. It features a user-friendly interface for editing and filtering data within the data table, as well as a detailed entity card for adding new entries or modifying existing ones.\\n\\nIn the JSON structure, each key represents a distinct data model, encompassing three primary components: \\"description,\\" \\"columns_json,\\" and \\"statistics_json\\"\\n\\n\\"columns_json\\": This component houses the data fields, each characterized by a type and a label. It's an object with key - name of field, value - it's properties. Valid types include \\"INTEGER\\", \\"DATE\\", \\"BOOLEAN\\", \\"FLOAT\\", \\"TEXT\\", and \\"STRING\\". The label is a text field specifying the name of the field. Label text must be in same language as users message. Fields can optionally include a \\"default\\" parameter indicating a preset value and a \\"unique\\" parameter (set to either \\"true\\" or \\"false\\") that dictates whether the field values should be distinct. Additionally, fields can contain an \\"association\\" object signifying a relationship with other models. This object comprises parameters such as \\"type\\" (indicating the relationship nature, either \\"one-to-one\\" or \\"one-to-many\\"), \\"model_name\\" (the name of the connected model), and \\"model_field\\" (the field used as the display value post-linkage). The key name of the field with an association should match the \\"model_field\\" with which it connects. The \\"sortable\\" parameter, which can be set to either \\"true\\" or \\"false\\", specifies whether the field can be sorted in the table. Other optional parameters that affect display properties include \\"display_type\\", \\"non_editable\\", \\"hidden_all\\", \\"click_action\\", \\"width\\", and \\"hidden_in_table\\".\\n\\n\\"statistics_json\\": This array defines the specifications for statistics charts related to the model. It includes properties such as:\\n\\"type\\": The type of chart (e.g., \\"bar\\").\\n\\"color\\": The default color for the chart.\\n\\"label\\": The chart's label.\\n\\"measure\\": The aggregation method (e.g., \\"sum\\" or \\"count\\").\\n\\"group_by\\": The field used for aggregation.\\n\\"sum_field\\": The specific field subject to aggregation in the sum or count.\\n\\nThrough this structured format, the \\"custom data\\" module streamlines the integration of custom-defined data into the bot's operational framework, enhancing its adaptability and functional range. Users message must describe one or many custom-data models. On output completion must be trimmed, no extra spaces, no extra line-breaks json-object of models even if it is one. Key must be model name, in english. No other text. And don't skip fields in JSON. It will be parsed. If you can't write json with custom data just output [] empty array."