pip3 install frelatage
Current release : 0.0.7
Frelatage is a coverage-based Python fuzzing library which can be used to fuzz python code. The development of Frelatage was inspired by various other fuzzers, including AFL/AFL++, Atheris and PythonFuzz. The main purpose of the project is to take advantage of the best features of these fuzzers and gather them together into a new tool in order to efficiently fuzz python applications.
DISCLAIMER : This project is at the alpha stage and can still cause many unexpected behaviors. Frelatage should not be used in a production environment at this time.
Install with pip (recommended)
pip3 install frelatage
Or build from source
Recommended for developers. It automatically clones the main branch from the frelatage repo, and installs from source.
# Automatically clone the Frelatage repository and install Frelatage from sourcebash <(wget -q https://raw.githubusercontent.com/Rog3rSm1th/Frelatage/main/scripts/autoinstall.sh -O -)
How it works
The idea behind the design of Frelatage is the usage of a genetic algorithm to generate mutations that will cover as much code as possible. The functioning of a fuzzing cycle can be roughly summarized with this diagram :
graph TB m1(Mutation 1) --> |input| function(Fuzzed function) m2(Mutation 2) --> |input| function(Fuzzed function) mplus(Mutation ...) --> |input| function(Fuzzed function) mn(Mutation n) --> |input| function(Fuzzed function) function --> generate_reports(Generate reports) generate_reports --> rank_reports(Rank reports) rank_reports --> select(Select n best reports) select --> |mutate| nm1(Mutation 1) & nm2(Mutation 2) & nmplus(Mutation ...) & nmn(Mutation n) subgraph Cycle mutations direction LR m1 m2 mplus mn end subgraph Next cycle mutations direction LR nm1 nm2 nmplus nmn end style function fill:#5388e8,stroke:white,stroke-width:4px
Fuzzing different argument types:
Frelatage allows to fuzz a function by passing a file as input.
Fuzz a classical parameter
import frelatageimport my_vulnerable_librarydef MyFunctionFuzz(data): my_vulnerable_library.parse(data)input = frelatage.Input(value="initial_value")f = frelatage.Fuzzer(MyFunctionFuzz, [[input]])f.fuzz()
Fuzz a file parameter
Frelatage gives you the possibility to fuzz file type input parameters. To initialize the value of these files, you must create files in the input folder (
./in by default).
If we want to initialize the value of a file used to fuzz, we can do it like this:
echo "initial value" > ./in/input.txt
And then run the fuzzer:
import frelatageimport my_vulnerable_librarydef MyFunctionFuzz(data): my_vulnerable_library.load_file(data)input = frelatage.Input(file=True, value="input.txt")f = frelatage.Fuzzer(MyFunctionFuzz, [[input]])f.fuzz()
Load several files to a corpus at once
If you need to load several files into a corpus at once (useful if you use a large corpus) You can use the built-in function of Frelatage
load_corpus. This function returns a list of inputs.
load_corpus(directory: str, file_extensions: list) -> list[Input]
- directory: Subdirectory of the input directory (relative path), e.g
- file_extensions: List of file extensions to include in the corpus entries, e.g.
import frelatageimport my_vulnerable_librarydef MyFunctionFuzz(data): my_vulnerable_library.load_file(data) my_vulnerable_library.load_file(data2)# Load every every file in the ./in directorycorpus_1 = frelatage.load_corpus(directory="./")# Load every .gif/.jpeg file in the ./in/images subdirectorycorpus_2 = frelatage.load_corpus(directory="./images", file_extension=["gif", "jpeg"])f = frelatage.Fuzzer(MyFunctionFuzz, [corpus_1, corpus_2])f.fuzz()
Fuzz with a dictionary
You can copy one or more dictionaries located here in the directory dedicated to dictionaries (
./dict by default).
Differental fuzzing is a popular software testing technique that attempts to detect bugs by providing the same input to multiple libraries/programs and observing differences in their behaviors. You will find an example here of a use of differential fuzzing with Frelatage with the
You can find more examples of fuzzers and corpus in the examples directory.
Each crash is saved in the output folder (
./out by default), in a folder named :
id:<crash ID>,err:<error type>,err_pos:<error>,err_file:<error file>.
The report directory is in the following form:
├── out │ ├── id:<crash ID>,err:<error type>,err_file:<error file>,err_pos:<err_pos> │ ├── input │ ├── 0 │ ├── <inputfile1> │ ├── ... │ ├── ...
Read a crash report
Inputs passed to a function are serialized using the pickle module before being saved in the
<report_folder>/input file. It is therefore necessary to deserialize it to be able to read the contents of the file. This action can be performed with this script.
There are two ways to set up Frelatage:
Using the environment variables
|ENV Variable||Description||Possible Values||Default Value|
|FRELATAGE_DICTIONARY_ENABLE||Enable the use of mutations based on dictionary elements|| || |
|FRELATAGE_TIMEOUT_DELAY||Delay in seconds after which a function will return a TimeoutError|| || |
|FRELATAGE_INPUT_FILE_TMP_DIR||Temporary folder where input files are stored||absolute path to a folder, e.g. || |
|FRELATAGE_INPUT_MAX_LEN||Maximum size of an input variable in bytes|| || |
|FRELATAGE_MAX_THREADS||Maximum number of simultaneous threads|| || |
|FRELATAGE_MAX_CYCLES_WITHOUT_NEW_PATHS||Number of cycles without new paths found after which we go to the next stage|| || |
|FRELATAGE_INPUT_DIR||Directory containing the initial input files. It needs to be a relative path (to the path of the fuzzing file)||relative path to a folder, e.g. || |
|FRELATAGE_DICTIONARY_DIR||Default directory for dictionaries. It needs to be a relative path (to the path of the fuzzing file)||relative path to a folder, e.g. || |
|FRELATAGE_DEBUG_MODE||Enable the debug mode (show the error when Frelatage crash)|| || |
A configuration example :
export FRELATAGE_DICTIONARY_ENABLE=1 &&export FRELATAGE_TIMEOUT_DELAY=2 &&export FRELATAGE_INPUT_FILE_TMP_DIR="/tmp/frelatage" &&export FRELATAGE_INPUT_MAX_LEN=4096 &&export FRELATAGE_MAX_THREADS=8 &&export FRELATAGE_MAX_CYCLES_WITHOUT_NEW_PATHS=5000 &&export FRELATAGE_INPUT_DIR="./in" &&export FRELATAGE_DICTIONARY_DIR="./dict" &&python3 fuzzer.py
Passing arguments to the fuzzer
import frelatage def myfunction(input1_string, input2_int): passinput1 = frelatage.Input(value="initial_value")input2 = frelatage.Input(value=2)f = frelatage.Fuzzer( # The method you want to fuzz method=myfunction, # Corpus corpus=[[input1], [input2]], # Number of threads threads_count=8, # Exceptions that will be taken into account exceptions_whitelist=(OSError), # Exceptions that will not be taken into account exceptions_blacklist=(), # Directory where the error reports will be stored output_directory="./out", # Enable or disable silent mode silent=False)f.fuzz()
Please keep in mind that, similarly to many other computationally-intensive tasks, fuzzing may put strain on your hardware and on the OS. In particular:
Your CPU will run hot and will need adequate cooling. In most cases, if cooling is insufficient or stops working properly, CPU speeds will be automatically throttled. That said, especially when fuzzing on less suitable hardware (laptops, smartphones, etc), it's not entirely impossible for something to blow up.
Targeted programs may end up erratically grabbing gigabytes of memory or filling up disk space with junk files. Frelatage tries to enforce basic memory limits, but can't prevent each and every possible mishap. The bottom line is that you shouldn't be fuzzing on systems where the prospect of data loss is not an acceptable risk.
Fuzzing involves billions of reads and writes to the filesystem. On modern systems, this will be usually heavily cached, resulting in fairly modest "physical" I/O - but there are many factors that may alter this equation. It is your responsibility to monitor for potential trouble; with very heavy I/O, the lifespan of many HDDs and SSDs may be reduced.
A good way to monitor disk I/O on Linux is the 'iostat' command:
$ iostat -d 3 -x -k [...optional disk ID...]
About Me/Hire me
I am Rog3rSm1th, I am 21 years old and I'm a French computer and cybersecurity enthusiast. I like developing tools (OSINT, Fuzzing...) and playing CTFs/Wargames. To learn more about me and my projects, juste click here.
➜ If you want to hire me for one of your projects (Programming, cybersecurity...), just contact me at [email protected] and we will assess your needs together.