https://github.com/gamozolabs/cookie_dough

A fuzzing introspection tool
https://github.com/gamozolabs/cookie_dough

Last synced: 22 days ago
JSON representation

A fuzzing introspection tool

• Host: GitHub
• URL: https://github.com/gamozolabs/cookie_dough
• Owner: gamozolabs
• License: mit
• Created: 2020-07-29T11:07:15.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2020-09-13T13:43:30.000Z (about 4 years ago)
• Last Synced: 2024-08-05T17:42:51.425Z (4 months ago)
• Language: Rust
• Size: 3.13 MB
• Stars: 160
• Watchers: 7
• Forks: 20
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-hacking-lists - gamozolabs/cookie_dough - A fuzzing introspection tool (Rust)

README

        
# Summary

This is a environment for measuring and observing the behaviors of fuzzers from

inside the fuzzer itself.

This allows for comparing fuzzers without having to worry about them having

different definitions of performance and coverage.

# Visualizations

![](_assets/example_data.gif)

This tool provides real-time visualizations of coverage as it happens, so you

can not only see what was hit, but what is being hit frequently!

It's important to note, while this tool currently mainly does visualizations,

we'll likely start working on analyizing results statistically rather than

eyeballing images!

# Usage

To use this, simply run `cargo run` in `prog_gen` to create a random program.

This random program expects a filename as the argument once compiled which is

then used as the fuzz input for the randomly generated program.

This program then reports status via shared memory to a file which can be

observed and monitored in a web browser by using `coverage_server`. Run `cargo

run` in coverage server and it will monitor the shared memory for changes on an

interval, and when new changes are observed the information will be pushed over

a websocket to a website which will visualize the coverage in realtime via

graphviz SVG.

## TL;DR

Run `cargo run` in `prog_gen` to generate a random program.

Run `cargo run` in `coverage_server` to monitor coverage events.

Run `python3 -m http.server` in `coverage_server` to host the coverage viewer

website, and navigate to the locally hosted website.