diff --git a/src/file.S b/src/cat.S
similarity index 61%
rename from src/file.S
rename to src/cat.S
index 76ccae6..99a25d3 100644
--- a/src/file.S
+++ b/src/cat.S
@@ -57,11 +57,31 @@ cat_file:
     la a2, mod_buf_0 # load mod_buf_0 as write buf
     2:
 
-    call _add_endings
+    call add_endings
     xori s1, s1, 1 # flip the mod buf to write to
     mv s2, a1 # update the current byte count
     .LNoEndings:
 
+    la t1, number
+    lb t1, 0(t1)
+    beqz t1, .LNoNumbers
+
+    mv a1, s2 # load mod buf size to a1
+
+    beqz s1, 1f
+    la a0, mod_buf_0 # load mod_buf_0 as read buf
+    la a2, mod_buf_1 # load mod_buf_1 as write buf
+    j 2f
+    1:
+    la a0, mod_buf_1 # load mod_buf_1 as read buf
+    la a2, mod_buf_0 # load mod_buf_0 as write buf
+    2:
+
+    call add_line_numbers
+    xori s1, s1, 1 # flip the mod buf to write to
+    mv s2, a1 # update the current byte count
+    .LNoNumbers:
+
     beqz s1, 1f 
     la a1, mod_buf_0
     j 2f
@@ -109,67 +129,3 @@ open_fd:
 
     2:
     ret
-
-# Print to stdout while adding "$" characters.
-# @param[in] a0: ptr to null-terminated str.
-# @param[in] a1: length of str.
-# @param[in] a2: mod buf.
-# @param[out] a0: result code.
-# @param[out] a1: bytes written to mod buf.
-_add_endings:
-    addi sp, sp, -72
-    sd s7, 64(sp)
-    sd s6, 56(sp)
-    sd s5, 48(sp)
-    sd s4, 40(sp)
-    sd ra, 32(sp)
-    sd s3, 24(sp)
-    sd s2, 16(sp)
-    sd s1, 8(sp)
-    sd s0, 0(sp)
-
-
-    mv s0, a0 # put the cat_buf ptr in s0
-    mv s1, a1
-    li s2, 0 # set the cat_buf char counter to 0
-    li s3, 0 # set the endings_buf char counter to 0
-    
-    mv s5, a2
-    .LStdoutEndingsProcessChunk:
-    add s4, s0, s2 # Calculate the offset address
-    lb t6, 0(s4)
-    li t4, 0xA # newline
-    bne t6, t4, 1f # if the current byte is not a newline, skip to writing
-
-    add s4, s5, s3 # s4 = endings_buf[s2]
-    li t5, '$'
-    sb t5, 0(s4) # endings_buf[s2] = $
-    li t5, '\n'
-    addi s4, s4, 1
-    sb t5, 0(s4) # endings_buf[s2] = \n
-    addi s3, s3, 2
-    j 2f
-
-    1:
-    add s4, s5, s3 # s4 = endings_buf[s2]
-    sb t6, 0(s4) # endings_buf[s2] = t6
-
-    addi s3, s3, 1 # inc endings_buf pointer
-    2:
-    addi s2, s2, 1 # inc cat_buf pointer
-    blt s2, s1, .LStdoutEndingsProcessChunk
-
-    li a0, 0
-    mv a1, s3
-
-    ld s0, 0(sp)
-    ld s1, 8(sp)
-    ld s2, 16(sp)
-    ld s3, 24(sp)
-    ld ra, 32(sp)
-    ld s4, 40(sp)
-    ld s5, 48(sp)
-    ld s6, 56(sp)
-    ld s7, 64(sp)
-    addi sp, sp, 72
-    ret
diff --git a/src/main.S b/src/main.S
index def4c09..13323f0 100644
--- a/src/main.S
+++ b/src/main.S
@@ -1,6 +1,7 @@
 .data
 argc:           .space 8
 show_ends:      .space 1
+number:         .space 1
 
 cat_prefix:     .ascii "cat: "
 .equ CAT_PREFIX_LEN, . - cat_prefix
@@ -13,13 +14,14 @@ help:
     .ascii "Concatenate FILE(S) to standard output.\n\n"
     .ascii "With no FILE, or when FILE is -, read standard input.\n\n"
     .ascii "    -E, --show-ends     display $ at end of each line\n"
+    .ascii "    -n, --number        number all output lines\n"
     .ascii "    --help              display this help and exit\n"
     .ascii "    --version           output version information and exit\n\n"
     .asciz "Git repository: <https://git.lexza.ch/Lexzach/riscv-cat>\n"
 .equ HELP_LEN, . - help
 
 version:
-    .ascii "cat 1.2.0\n"
+    .ascii "cat 1.3.0\n"
     .ascii "Copyright (C) 2026 lexzach\n"
     .ascii "License MIT\n"
     .ascii "This is free software: you are free to change and redistribute it.\n"
@@ -31,9 +33,11 @@ flag_help:     .asciz "--help"
 flag_version:  .asciz "--version"
 flag_show_ends:  .asciz "--show-ends"
 flag_E:  .asciz "-E"
+flag_number: .asciz "--number"
+flag_n: .asciz "-n"
 
 .section .text
-.globl _start, show_ends
+.globl _start, show_ends, number
 
 # main.
 _start:
@@ -48,6 +52,8 @@ _start:
     li s1, 1
     li s2, 0 # processed files
     
+    call zero_line_count
+    bnez a0, .LMainReturnErr
     call _check_flags
 
     li t0, 1
@@ -150,6 +156,12 @@ _check_flags:
     bnez a0, .LMainReturnErr
     sb a1, show_ends, t1 # since the return for has switch is either 0 for not found or 1 for found, we can just write it directly into the buf
     
+    la a0, flag_number
+    la a1, flag_n
+    call has_switch
+    bnez a0, .LMainReturnErr
+    sb a1, number, t1
+
     ld ra, 0(sp)
     addi sp, sp, 8
     ret
@@ -185,6 +197,16 @@ _is_known_flag:
     call str_cmp
     bnez a1, .LFlagFound
 
+    la a0, flag_number
+    mv a1, s0
+    call str_cmp
+    bnez a1, .LFlagFound
+
+    la a0, flag_n
+    mv a1, s0
+    call str_cmp
+    bnez a1, .LFlagFound
+
     li a1, 0
     .LFlagFound:
     ld s0, 0(sp)
diff --git a/src/mod.S b/src/mod.S
new file mode 100644
index 0000000..19fa675
--- /dev/null
+++ b/src/mod.S
@@ -0,0 +1,208 @@
+.data
+line_count:     .space 6
+
+.section .text
+.globl add_endings, add_line_numbers, zero_line_count
+
+# Print to stdout while adding "$" characters.
+# @param[in] a0: ptr to null-terminated str.
+# @param[in] a1: length of str.
+# @param[in] a2: mod buf addr.
+# @param[out] a0: result code.
+# @param[out] a1: bytes written to mod buf.
+add_endings:
+    addi sp, sp, -72
+    sd s7, 64(sp)
+    sd s6, 56(sp)
+    sd s5, 48(sp)
+    sd s4, 40(sp)
+    sd ra, 32(sp)
+    sd s3, 24(sp)
+    sd s2, 16(sp)
+    sd s1, 8(sp)
+    sd s0, 0(sp)
+
+
+    mv s0, a0 # put the str buf ptr in s0
+    mv s1, a1
+    li s2, 0 # set the str buf char counter to 0
+    li s3, 0 # set the mod buf char counter to 0
+    
+    mv s5, a2
+    .LEndingsProcessChunk:
+    add s4, s0, s2 # Calculate the offset address
+    lb t6, 0(s4)
+    li t4, 0xA # newline
+    bne t6, t4, 1f # if the current byte is not a newline, skip to writing
+
+    add s4, s5, s3 # s4 = mod buf[s2]
+    li t5, '$'
+    sb t5, 0(s4) # mod buf[s2] = $
+    li t5, '\n'
+    addi s4, s4, 1
+    sb t5, 0(s4) # mod buf[s2] = \n
+    addi s3, s3, 2
+    j 2f
+
+    1:
+    add s4, s5, s3 # s4 = mod buf[s2]
+    sb t6, 0(s4) # mod buf[s2] = t6
+
+    addi s3, s3, 1 # inc mod buf pointer
+    2:
+    addi s2, s2, 1 # inc str buf pointer
+    blt s2, s1, .LEndingsProcessChunk
+
+    li a0, 0
+    mv a1, s3
+
+    ld s0, 0(sp)
+    ld s1, 8(sp)
+    ld s2, 16(sp)
+    ld s3, 24(sp)
+    ld ra, 32(sp)
+    ld s4, 40(sp)
+    ld s5, 48(sp)
+    ld s6, 56(sp)
+    ld s7, 64(sp)
+    addi sp, sp, 72
+    ret
+
+# Prefix newlines with ######+TAB line count
+# @param[in] a0: ptr to null-terminated str.
+# @param[in] a1: length of str.
+# @param[in] a2: mod buf addr.
+# @param[in] a3: skip empty.
+# @param[out] a0: result code.
+# @param[out] a1: bytes written to mod buf.
+add_line_numbers:
+    addi sp, sp, -72
+    sd s7, 64(sp) #unused
+    sd s6, 56(sp) # mod buf ptr
+    sd s5, 48(sp) # mod buf bp
+    sd s4, 40(sp) # str buf ptr
+    sd ra, 32(sp)
+    sd s3, 24(sp) # mod buf offset
+    sd s2, 16(sp) # str buf offset
+    sd s1, 8(sp) # len of str
+    sd s0, 0(sp) # str buf addr
+
+
+    mv s0, a0 # put the str ptr in s0
+    mv s1, a1 # move length of str into s1
+    li s2, 0 # set the str buf char offset to 0
+    li s3, 0 # set the mod buf char offset to 0
+    mv s5, a2 # move mod buf addr into s5
+    
+    .LLineNumbersProcessChunk:
+    add s4, s0, s2 # Calculate the offset address
+    lb t6, 0(s4)
+    li t4, 0xA # newline
+    bne t6, t4, 1f # if the current byte is not a newline, skip to writing
+    
+    call _inc_line_count
+    
+    addi s3, s3, 1 # we want the ptr at the spot right after the newline
+    add s6, s5, s3 # update mod buf + offset addr
+    
+    li t2, 0 # lc char buf offset
+    li t3, 6 # lc char buf max offset
+    la t1, line_count # load the line count buffer addr
+    li t5, 0xA # newline character
+    sb t5, 0(s6)
+    addi s3, s3, 1
+    add s6, s5, s3 # update mod buf + offset addr
+    3:
+    add t4, t1, t2
+    lb t5, 0(t4)
+    sb t5, 0(s6) # mod buf[s4] = t4
+    addi t2, t2, 1
+    addi s3, s3, 1
+    add s6, s5, s3 # update mod buf + offset addr
+    blt t2, t3, 3b
+
+    li t5, 0x7 # tab character
+    sb t5, 0(s6)
+
+    addi s3, s3, 1
+    
+    j 2f
+
+    1:
+    add s6, s5, s3 # s4 = mod buf[s2]
+    sb t6, 0(s6) # mod buf[s2] = t6
+
+    addi s3, s3, 1 # inc mod buf pointer
+    2:
+    addi s2, s2, 1 # inc str buf pointer
+    blt s2, s1, .LLineNumbersProcessChunk
+
+    li a0, 0
+    mv a1, s3
+
+    ld s0, 0(sp)
+    ld s1, 8(sp)
+    ld s2, 16(sp)
+    ld s3, 24(sp)
+    ld ra, 32(sp)
+    ld s4, 40(sp)
+    ld s5, 48(sp)
+    ld s6, 56(sp)
+    ld s7, 64(sp)
+    addi sp, sp, 72
+    ret
+
+
+# Increment the line count string
+# @param[in] a0: not used
+# @param[out] a0: result code
+_inc_line_count:
+    la t0, line_count # load the line count buffer address
+    li t1, 5 # current offset, we move from right to left first
+
+    .LIncLine:
+    bltz t1, .LLineCountDoneInc
+    add t2, t0, t1 # load the correct offset address into t2
+    lb t3, 0(t2) # load the current ascii character into t3
+    
+    li t4, 32 # load space character
+    bne t3, t4, .LLineCountNotSpaceChar # if the character is not a space, skip setting to zero
+    li t4, 48 # ascii '0'
+    sb t4, 0(t2) # store this at the current pointer
+
+    .LLineCountNotSpaceChar:
+    add t2, t0, t1 # load the correct offset address into t2
+    lb t3, 0(t2) # load the current ascii character into t3
+    li t4, 58 # ascii ':', right after ascii '9', overflow
+    
+    addi t3, t3, 1
+    beq t3, t4, .LLineCountOverflow
+    sb t3, 0(t2)
+    j .LLineCountDoneInc
+    .LLineCountOverflow:
+    li t4, 48 # ascii '0'
+    sb t4, 0(t2) # store that ascii '0'
+    li t4, 1
+    sub t1, t1, t4 # move to the next digit to the left
+    j .LIncLine
+
+    .LLineCountDoneInc:
+    ret
+
+# Set the line count to all space characters
+# @param[in] a0: not used
+# @param[out] a0: result code
+zero_line_count:
+    la t0, line_count # load the line count buffer address
+    li t1, 0 # char ptr
+    li t2, 6 # max char
+    li t3, 32 # space character
+
+    1:
+    add t4, t0, t1 # load offset addr
+    sb t3, 0(t4) # store the space char in offset addr
+    addi t1, t1, 1
+    blt t1, t2, 1b
+
+    li a0, 0
+    ret