working with List of Maps

You can use group by group or try using where contains function.

You could use the reduce() function to iterate on the input data and create a output map. Some ideas here: https://appian.rocks/2022/08/29/complex-algorithms-in-appian/

Below some example code that picks unique random values from a list. This pattern allows you to put all the code into a single expression. This code does not implement a solution to your use case!

if(
  ri!isRecursion,
  a!localVariables(
    local!index: ceiling(rand() * count(ri!data.chars)),
    a!map(
      chars: remove(ri!data.chars, local!index),
      output: append(ri!data.output, ri!data.chars[local!index])
    )
  ),
  reduce(
    rule!SSH_RandomPicker(
      isRecursion: true,
      data:_,
      num:_
    ),
    a!map(
      chars: char(65 + enumerate(26)),
      output: {}
    ),
    enumerate(10)
  )
)

It doesn't even really require a nested ForEach or (if you'd rather avoid the legacy function) reduce(), just a bit of reassembly using a!forEach().

a!localVariables(
  local!Grades: {
    a!map(Group: "A", Excellent: 0, Good: 0, Regular: 0),
    a!map(Group: "A", Excellent: 2, Good: 0, Regular: 0),
    a!map(Group: "B", Excellent: 0, Good: 0, Regular: 2),
    a!map(Group: "B", Excellent: 0, Good: 2, Regular: 0),
    a!map(Group: "C", Excellent: 0, Good: 2, Regular: 0),
  },
  
  local!uniqueGroups: rule!RULE_General_distinct(    
        /* simple helper rule that unions an array against itself */
    value: local!Grades.Group
  ),
  
  a!forEach(
    local!uniqueGroups,
    
    a!localVariables(
      local!listForCurrentGroup: index(
        local!Grades,
        wherecontains(fv!item, local!Grades.Group),
        {}
      ),
      a!map(
        Group: fv!item,
        ExcellentSum: sum(local!listForCurrentGroup.Excellent),
        GoodSum: sum(local!listForCurrentGroup.Good),
        RegularSum: sum(local!listForCurrentGroup.Regular)
      )
    )
  )
)

Thank you for your replied!

Thank you for your replied!