BASH: Script to process VEGA output files

#!/bin/bash

# Function to remove lines with [Error]
remove_error_lines() {
  sed '/\[Error\]/d' "$1" > "$1.tmp" && mv "$1.tmp" "$1"
}

# Process each tab-delimited file
merged_files=""
for file in *.tsv *.txt; do # Adjust file extensions if needed
  if [ -f "$file" ]; then
    tmp_file="${file}.processed"

    # Remove first 10 lines
    if [[ "$OSTYPE" == "darwin"* ]]; then
      sed '1,10d' "$file" > "$tmp_file"
    else
      sed '1,10d' "$file" > "$tmp_file"
    fi

    # Append to merged_files
    merged_files+=" $tmp_file"
  fi
done

# Merge files
cat $merged_files > merged.txt

# Remove first 2 columns and [Error] lines from merged.txt
awk '{ $1=$2=""; sub(/^  */,""); print }' merged.txt > merged.tmp
remove_error_lines merged.tmp
mv merged.tmp merged.txt

# Add incrementing first column
awk 'NR==1{print} NR>1{print NR-1 "\t" $0}' merged.txt > merged_final.txt
mv merged.txt merged_old.txt
mv merged_final.txt merged.txt

# Cleanup
rm *.processed *.tmp merged_old.txt

echo "Processing complete. Output in merged.txt"

# Gemini prompt:

Provide a bash script that wroks on MacOS and Linux that processes all tab-delimited files in a 
directory as follows: delete the first 10 lines of each file, then merge the files into a single 
file called 'merged.txt'.

In this file remove the first 2 columns. Then delete any lines that contain the text: [Error].
Finally, add a new first column and fill it with incrementing numbers starting at 1. 
Make sure that each file is not processed in-place.