Skip to content

Forensic framework to build tools that can be reused in multiple projects without changing anything

License

Notifications You must be signed in to change notification settings

ForensicRS/forensic-rs

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

53 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Forensic-rs

crates.io documentation MIT License Rust

A Rust-based framework to build tools that analyze forensic artifacts and can be reused as libraries across multiple projects without changing anything.

Note: still in Alpha version

Community

Discord

Join the conversation on Discord to discuss anything related to ForensicRS.

Introduction

The idea behind the framework is to allow the reuse of forensic artifact analysis tools. For this reason, the framework decouples the code of the analysis tools that parses or reads artifacts from the ones that access the readed value: registry keys, files, SQL rows. Thus, a tool that analyzes UAL artifacts can be used regardless of whether the artifact is inside a ZIP as a result of triage or directly on the file system.

In this way, the same tools can be used if we want to make a triage processor like Plaso, a module within an EDR or even a tool with a graphical interface like Eric Zimmerman's Registry Explorer with the advantage of the reliability of the Rust code and its easy integration into Python scripts.

Supported artifacts

  • Windows Registry: See RegistryReader trait.
  • SQL databases: See SqlStatement trait. There is also a basic wrapper example around the sqlite crate in sql_tests.
  • File Systems: With this trait we can read files and directories. It is very useful because we can stack file systems: A file inside a OleObject inside a ZIP file that is also inside a ZIP. See VirtualFileSystem and the implementation using the standard library (std::fs) in StdVirtualFS.

Registry Example

So in this framework we will have libraries that allows us to access the Windows registry. One in a live environment using the Windows API, and another one that parses a registry hive. So we will also have libraries that extracts data from the registry, theses libraries need to be decoupled from the registry access implementation.

Here is where this framework comes to help with the traits:

pub trait RegistryReader {
    fn open_key(&mut self, hkey : RegHiveKey, key_name : &str) -> ForensicResult<RegHiveKey>;
    fn read_value(&self, hkey : RegHiveKey, value_name : &str) -> ForensicResult<RegValue>;
    fn enumerate_values(&self, hkey : RegHiveKey) -> ForensicResult<Vec<String>>;
    fn enumerate_keys(&self, hkey : RegHiveKey) -> ForensicResult<Vec<String>>;
    fn key_at(&self, hkey : RegHiveKey, pos : u32) -> ForensicResult<String>;
    fn value_at(&self, hkey : RegHiveKey, pos : u32) -> ForensicResult<String>;
}

So now we can write our analysis library without knowing if we are accessing a live system or a hive file.

  • LiveRegistry Library: implements the RegistryReader trait.
  • HiveParser Library: implements the RegistryReader trait.
  • ShellBags analyzer: accepts a RegistryReader as a parameter to access the registry.

And ShellBags analyzer can be used in a EDR-like agent or as a analysis tool in a forensic case.

SQL Example

Extracted from the SQL trait tests using sqlite db.

let conn = prepare_db();
let w_conn = prepare_wrapper(conn);
let mut statement = w_conn.prepare("SELECT name, age FROM users;").unwrap();
test_database_content(statement.as_mut()).expect("Should not return error");

fn test_database_content<'a>(statement : &mut dyn SqlStatement) -> ForensicResult<()> {
    assert!(statement.next()?);
    let name : String = statement.read(0)?.try_into()?;
    let age : usize = statement.read(1)?.try_into()?;
    assert_eq!("Alice", name);
    assert_eq!(42, age);
    assert!(statement.next()?);
    let name : String = statement.read(0)?.try_into()?;
    let age : usize = statement.read(1)?.try_into()?;
    assert_eq!("Bob", name);
    assert_eq!(69, age);
    assert!(!statement.next()?);
    Ok(())
}

VFS Example

Extracted from StdVirtualFS tests using sqlite db.

const CONTENT: &'static str = "File_Content_Of_VFS";
let tmp = std::env::temp_dir();
let tmp_file = tmp.join("test_vfs_file.txt");
let mut file = std::fs::File::create(&tmp_file).unwrap();
file.write_all(CONTENT.as_bytes()).unwrap();
drop(file);

let std_vfs = StdVirtualFS::new();
test_file_content(&std_vfs,&tmp_file);

fn test_file_content(std_vfs : &impl VirtualFileSystem, tmp_file : &PathBuf) {
    let content = std_vfs.read_to_string(tmp_file).unwrap();
    assert_eq!(CONTENT, content);
    
}

Logs

To simplify the development of modules, plugins and libraries its availabe some macros with the same syntax as that of the log crate:

// For production use initialize_logger(logger) instead of testing_logger_dummy()
let log_receiver = testing_logger_dummy();
error!("This is log name: {}", "ERROR");
warn!("This is log name: {}", "WARN");
info!("This is log name: {}", "INFO");
debug!("This is log name: {}", "DEBUG");
trace!("This is log name: {}", "TRACE");
assert_eq!("This is log name: ERROR", log_receiver.recv().unwrap());

Notifications and Alerts

To simplify the detection of anomalies when processing or analyzing artifacts, we can use the notifications. It uses a syntax similar as that of the log crate.

// For production use initialize_notifier(notifier) instead of testing_notifier_dummy()
let notification_receiver = testing_notifier_dummy();
notify_high!(NotificationType::AntiForensicsDetected, "The registry key {} is not present. The only possibility is that someone deleted it.", r"HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\ProfileList");
assert_eq!(r"The registry key HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\ProfileList is not present. The only possibility is that someone deleted it.", notification_receiver.recv().unwrap().data);

List of libraries

About

Forensic framework to build tools that can be reused in multiple projects without changing anything

Topics

Resources

License

Code of conduct

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages