You can read the entire contents of CSV and TXT files using the aptly named CSV/TXT package.

Most of the time, you don’t need to initialize a data table variable, the actions in which their output is a data table change the schema of the table: i.e.: the headers, number of columns, number of rows, etc.

Here’s an example:

If your TXT file looks like this:

Department
Finance
IT
Human Resources
Sales

You could create a task bot like this one:

And the contents of the output data table variable would look like this: 

However, if your table variable does not follow the same schema as the source TXT/CSV, you can set up a different schema for that variable in the Variables panel by clicking the “Edit table (1x1)” button.

When you click that button, a dialog will come up where you can specify the name and data type of the columns as well as the number of rows and columns.

You can also do an in-line initialization with the Data Table: Assign action. That action also contains the “Edit table (1x1)” button, and it works just like the one in the Variables panel.

As for your second question, there are a couple different ways to add an element to a table variable, each with its own pros and cons, so it really depends on your use case.

These are some of the simplest ways to set the values of a table: 

• Use the table indexes / row index + column name: Set value of a single cell action (automationanywhere.com).  

• Add the values to a record variable and then insert the record variable to a data table: Update column action in the Record package (automationanywhere.com) + Insert row action (automationanywhere.com)

And again, these are not the only 2 ways, you may find more in this very forum as it is a common question.

How to assign value in the Data Table assign action? | Community (automationanywhere.com)

Personally, whenever I need to update a data table variable, I do something like this: