refactor(bare): enhance security and performance optimizations

- Remove unnecessary sorting in advertise_refs for deterministic output
- Add path traversal detection and validation in bare_dir construction
- Implement symlink resolution checks to prevent security vulnerabilities
- Refactor cache system with CRC validation and improved metrics
- Integrate repo-specific cache invalidation using indexed keys
- Add comprehensive unit tests for commit operations and diff functionality
- Move configuration constants to centralized config module
- Optimize string operations in disk cache random value generation
- Enhance license detection algorithm with cleaner matching logic
- Streamline argument processing in various git operations
- Update dependencies including crc32fast and flate2 for performance
- Add signal handling capability to tokio runtime configuration

hooks/sanitize.rs

@@ -25,7 +25,6 @@ const FORBIDDEN_PATTERNS: &[&str] = &[
     "init 6",
     "poweroff",
     "halt",
-    // Additional patterns to catch encoding/obfuscation attempts
     "eval ",     // eval can execute arbitrary strings
     "exec ",     // exec can replace process
     "$(",        // command substitution
@@ -55,8 +54,21 @@ const DANGEROUS_PREFIXES: &[&str] = &[
     "rm -rf *", // rm -rf with wildcard
 ];
 
-/// Maximum hook script size (64KB).
-const MAX_HOOK_SIZE: usize = 65536;
+/// Pairs of commands that indicate data exfiltration or code execution.
+const DANGEROUS_COMMAND_PAIRS: &[(&str, &str)] = &[
+    ("curl", "bash"),
+    ("curl", "sh"),
+    ("wget", "bash"),
+    ("wget", "sh"),
+    ("nc", "-e"),
+    ("ncat", "-e"),
+    ("python", "-c"),
+    ("perl", "-e"),
+    ("ruby", "-e"),
+    ("node", "-e"),
+];
+
+use crate::config::MAX_HOOK_SCRIPT_SIZE;
 
 /// Validate a custom hook script content for safety.
 pub fn validate_hook_content(content: &str) -> GitResult<()> {
@@ -65,10 +77,10 @@ pub fn validate_hook_content(content: &str) -> GitResult<()> {
             "hook content cannot be empty".into(),
         ));
     }
-    if content.len() > MAX_HOOK_SIZE {
+    if content.len() > MAX_HOOK_SCRIPT_SIZE {
         return Err(GitError::InvalidArgument(format!(
             "hook content too large (max {} bytes): {} bytes",
-            MAX_HOOK_SIZE,
+            MAX_HOOK_SCRIPT_SIZE,
             content.len()
         )));
     }
@@ -78,7 +90,6 @@ pub fn validate_hook_content(content: &str) -> GitResult<()> {
         ));
     }
 
-    // Check for forbidden patterns (case-insensitive where appropriate)
     let content_lower = content.to_lowercase();
     for pattern in FORBIDDEN_PATTERNS {
         if content_lower.contains(&pattern.to_lowercase()) {
@@ -88,7 +99,6 @@ pub fn validate_hook_content(content: &str) -> GitResult<()> {
         }
     }
 
-    // Check for dangerous prefixes (exact case)
     for prefix in DANGEROUS_PREFIXES {
         if content.contains(prefix) {
             return Err(GitError::InvalidArgument(format!(
@@ -97,15 +107,28 @@ pub fn validate_hook_content(content: &str) -> GitResult<()> {
         }
     }
 
-    // Check for obfuscation techniques
     check_obfuscation_attempts(content)?;
 
+    check_dangerous_pairs(content)?;
+
+    Ok(())
+}
+
+/// Check for dangerous command pairs that indicate data exfiltration or code execution.
+fn check_dangerous_pairs(content: &str) -> GitResult<()> {
+    let content_lower = content.to_lowercase();
+    for &(cmd1, cmd2) in DANGEROUS_COMMAND_PAIRS {
+        if content_lower.contains(cmd1) && content_lower.contains(cmd2) {
+            return Err(GitError::InvalidArgument(format!(
+                "hook contains dangerous command combination: '{cmd1}' + '{cmd2}' (possible data exfiltration)"
+            )));
+        }
+    }
     Ok(())
 }
 
 /// Check for common obfuscation attempts.
 fn check_obfuscation_attempts(content: &str) -> GitResult<()> {
-    // Check for excessive use of special characters that might indicate obfuscation
     let special_char_count = content
         .chars()
         .filter(|c| {
@@ -117,14 +140,12 @@ fn check_obfuscation_attempts(content: &str) -> GitResult<()> {
         .count();
     let total_chars = content.chars().count();
 
-    // If more than 30% of content is special characters, it's suspicious
     if total_chars > 0 && (special_char_count * 100 / total_chars) > 30 {
         return Err(GitError::InvalidArgument(
             "hook content appears obfuscated (too many special characters)".into(),
         ));
     }
 
-    // Check for hex encoding attempts (e.g., \x41\x42)
     if content.contains("\\x") {
         let hex_count = content.matches("\\x").count();
         if hex_count > 5 {
@@ -134,7 +155,6 @@ fn check_obfuscation_attempts(content: &str) -> GitResult<()> {
         }
     }
 
-    // Check for unicode escape sequences
     if content.contains("\\u") {
         let unicode_count = content.matches("\\u").count();
         if unicode_count > 5 {