SQLAlchemy Create & Update PostgreSQL variables, return w/Pandas

Question

I'm trying to create a Postgres script that needs to declare/update variables, return in a Pandas DataFrame.

This code executes:

conn = db_connection_object('sqlalchemy')
sql_query= '''  WITH my_cte as (
                select obs1  
                from my_data_table)
                SELECT * FROM my_cte
                '''
df = pd.read_sql_query(sql=sql_query,con=conn)
print(df)

This code crashes (notice DECLARE x integer;) :

conn = db_connection_object('sqlalchemy')
sql_query= '''  
                DECLARE x integer;
                WITH my_cte as (
                select obs1  
                from my_data_table)
                SELECT * FROM my_cte
                '''
df = pd.read_sql_query(sql=sql_query,con=conn)
print(df)

A simplified example of final use case (I know variables unneeded for this result, only to illustrate):

conn = db_connection_object('sqlalchemy')
sql_query= '''  
                DECLARE x integer;
                        y float8;
                WITH my_cte as (
                select obs1  
                from my_data_table)
                
                x = select count(obs1) FROM my_cte
                y = select avg(obs1) FROM my_cte
                SELECT x, y
                  '''
df = pd.read_sql_query(sql=sql_query,con=conn)
print(df)

I've read the PostgreSQL documentation, and several questions related to using pgAdmin4 and DBeaver as a query writing tools. I tried using pgAdmin4/ DBeaver for query creation and testing but they have different ways of handling variables and I can't tell if either syntax will work in my end use case.

Obviously I'd like to know how to accomplish the main question, and if anyone has a suggestion of a tool that uses the same syntax as my final use I'd be thankful to know :)

Thank you for helping me!