1#!/bin/bash
 2
 3# Script to check for hidden/invisible characters in Go files
 4# This helps detect potential prompt injection attempts
 5
 6echo "Checking Go files for hidden characters..."
 7
 8# Find all Go files in the repository
 9go_files=$(find . -name "*.go" -type f)
10
11# Counter for files with hidden characters
12files_with_hidden=0
13
14for file in $go_files; do
15  # Check for specific Unicode hidden characters that could be used for prompt injection
16  # This excludes normal whitespace like tabs and newlines
17  # Looking for:
18  # - Zero-width spaces (U+200B)
19  # - Zero-width non-joiners (U+200C)
20  # - Zero-width joiners (U+200D)
21  # - Left-to-right/right-to-left marks (U+200E, U+200F)
22  # - Bidirectional overrides (U+202A-U+202E)
23  # - Byte order mark (U+FEFF)
24  if hexdump -C "$file" | grep -E 'e2 80 8b|e2 80 8c|e2 80 8d|e2 80 8e|e2 80 8f|e2 80 aa|e2 80 ab|e2 80 ac|e2 80 ad|e2 80 ae|ef bb bf' > /dev/null 2>&1; then
25    echo "Hidden characters found in: $file"
26    
27    # Show the file with potential issues
28    echo "  Hexdump showing suspicious characters:"
29    hexdump -C "$file" | grep -E 'e2 80 8b|e2 80 8c|e2 80 8d|e2 80 8e|e2 80 8f|e2 80 aa|e2 80 ab|e2 80 ac|e2 80 ad|e2 80 ae|ef bb bf' | head -10
30    
31    files_with_hidden=$((files_with_hidden + 1))
32  fi
33done
34
35if [ $files_with_hidden -eq 0 ]; then
36  echo "No hidden characters found in any Go files."
37else
38  echo "Found hidden characters in $files_with_hidden Go file(s)."
39fi
40
41exit $files_with_hidden  # Exit with number of affected files as status code