Knowledge Management

Manage Knowledge Collections

The TaskingAI Python SDK provides a convenient way to manage your existing knowledge collections.

Before managing records using the SDK, you need to create a collection in the corresponding project on the TaskingAI platform. A collection is a container for storing records.

To obtain the collection_id, navigate to the management page of the created collection in your project. You can find the collection_id in the URL. For example, in the URL https://app.tasking.ai/projects/pX******/collections/DbgYX5Bq8B4R3BUg********/records, the collection_id is DbgYX5Bq8B4R3BUg********.

Collection Types

In TaskingAI, knowledge collections have two types: text and qa. In a text collection, records can be created and maintained through raw text, website URLs, or by directly uploading files. In a QA collection, records are created and maintained by uploading Q&A sheets in .csv or .xlsx format.

Before using related methods, make sure to confirm the type of collection and select the appropriate method for managing the records.

Manage Knowledge Records

In TaskingAI, records are the basic units to organize knowledge content in a collection. You can manage records by adding, updating, deleting, and querying them.

Each record has a unique record_id that can be used to identify and manage the record. To obtain the record_id, navigate to the collection management page in the TaskingAI platform, click the "more" button on the record you want to manage, and select "Copy ID" to copy the record_id.