ziggy

KO | EN

gitlite — search

#javascript #python #hacktoberfest #react #ai #llm #golang #rust #deep-learning #nodejs #awesome #awesome-list #pytorch #security

(★ 86)

A multi-fuzzer management utility for all of your Rust fuzzing needs 🧑‍🎤

ziggy Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# `ziggy`

[![Build status](https://github.com/srlabs/ziggy/actions/workflows/ci.yml/badge.svg)](https://github.com/srlabs/ziggy/actions/workflows/ci.yml)
[![Crates.io](https://img.shields.io/crates/v/ziggy.svg)](https://crates.io/crates/ziggy)
[![Docs.rs](https://img.shields.io/docsrs/ziggy)](https://docs.rs/ziggy)

`ziggy` is a fuzzer manager for Rust projects which is built to:

- launch different fuzzers in parallel with a shared corpus
- create and monitor continuous fuzzing pipelines

## Feature set

- 🤹 handling of different fuzzing processes in parallel ([honggfuzz](https://github.com/google/honggfuzz), [AFL++](https://github.com/aflplusplus/aflplusplus))
- 🗃️ one shared corpus for all fuzzers
- 🤏 effortless corpus minimization
- 📊 insightful monitoring
- 🎯 easy coverage report generation
- 😶‍🌫️ Arbitrary trait support

> Note: Fuzzing features such as CMPLOG or ASAN require ziggy to be built with the Rust nightly toolchain.

Features will also include:

- 🐇 [LibAFL](https://github.com/aflplusplus/libafl) integration
- 📨 notification of new crashes via bash hook

## Usage example

First, install `ziggy` and its dependencies by running:

```bash
cargo install --force ziggy cargo-afl honggfuzz
```

Here is the output of the tool's help:

```text
$ cargo ziggy
A multi-fuzzer management utility for all of your Rust fuzzing needs 🧑‍🎤

Usage: cargo ziggy <COMMAND>

Commands:
  build      Build the fuzzer and the runner binaries
  fuzz       Fuzz targets using different fuzzers in parallel
  run        Run a specific input or a directory of inputs to analyze backtrace
  minimize   Minimize the input corpus using the given fuzzing target
  cover      Generate code coverage information using the existing corpus
  plot       Plot AFL++ data using afl-plot
  add-seeds  Add seeds to the running AFL++ fuzzers
  triage     Triage crashes found with CASR - currently only works for AFL++
  stability  Analyze harness stability by detecting non-deterministic code paths
  clean      Remove generated artifacts from the target directory
  help       Print this message or the help of the given subcommand(s)

Options:
  -h, --help     Print help
  -V, --version  Print version
```

To create a fuzzer, simply add `ziggy` as a dependency.

```toml
[dependencies]
ziggy = { version = "1.7.2", default-features = false }
```

Then use the `fuzz!` macro inside your `main` to create a harness.

```rust ignore
fn main() {
    ziggy::fuzz!(|data: &[u8]| {
        println!("{data:?}");
    });
}
```

For a well-documented fuzzer, see [the url example](./examples/url/).

## The `output` directory

After you've launched your fuzzer, you'll find a couple of items in the `output/<target>` directory:

- the `corpus` directory containing the full corpus
- the `crashes` directory containing any crashes detected by the fuzzers
- the `timeouts` directory containing any timeouts/hangs detected by the fuzzers
- the `logs` directory containing fuzzer log files
- the `afl` directory containing AFL++'s output
- the `honggfuzz` directory containing Honggfuzz's output
- the `queue` directory that is used by ziggy to pass items from AFL++ to Honggfuzz

## Coverage

Generate an HTML coverage report from your existing shared corpus:

```bash
cargo ziggy cover
```

The report entry point will be in ziggy's `output` directory at `./<target>/coverage/index.html`.
It can be changed to `<dir>/index.html` using the `-o <dir>` option.

You can select different output formats or specify a custom input corpus:

```bash
cargo ziggy cover \
    -i path/to/corpus \  # custom input corpus directory
    -t html \            # navigable source-level report (default)
    -t text \            # same folder structure as html, but plain-text summaries
    -t json \            # machine-readable per-file data
    -t lcov              # standard tracefile for CI tools like Codecov
```

## Trophy case

[CVE-2026-24116](https://www.cve.org/CVERecord?id=CVE-2026-24116) was found in wasmtime by differential fuzzing with wasmi

ziggy

ziggy Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

ziggy

ziggy Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README